Kernel LLVM

AMD Phenom II X4 965 testing with a MSI 890FXA-GD70 (MS-7640) v1.0 (V1.8 BIOS) and MSI NVIDIA GeForce GTX 650 Ti 1GB on ManjaroLinux 21.0.4 via the Phoronix Test Suite.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2105120-HA-2105128HA60
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results

Limit displaying results to tests within:

C/C++ Compiler Tests 5 Tests
CPU Massive 9 Tests
Database Test Suite 4 Tests
Common Kernel Benchmarks 21 Tests
Memory Test Suite 3 Tests
Server 6 Tests
Server CPU Tests 4 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Hide
Result
Result
Identifier
Performance Per
Dollar
Date
Run
  Test
  Duration
Kernel LLVM
May 09 2021
  6 Hours, 27 Minutes
Kernel GCC
May 10 2021
  5 Hours, 21 Minutes
Kernel LLVM 11.1
May 10 2021
  6 Hours, 2 Minutes
Kernel GCC 11.1.1
May 12 2021
  8 Hours, 23 Minutes
Invert Hiding All Results Option
  6 Hours, 33 Minutes

Only show results where is faster than
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):

Kernel LLVM

Processor: AMD Phenom II X4 965 @ 3.30GHz (4 Cores), Motherboard: MSI 890FXA-GD70 (MS-7640) v1.0 (V1.8 BIOS), Chipset: AMD RD890 PCI-e GFX Hydra part, Memory: 16GB, Disk: 116GB Samsung SSD 840 + 200GB Seagate ST3200822AS, Graphics: MSI NVIDIA GeForce GTX 650 Ti 1GB, Audio: Realtek ALC892, Network: Realtek RTL8111/8168/8411

OS: ManjaroLinux 21.0.4, Kernel: 5.12.2-153-tkg-pds-llvm (x86_64), Desktop: KDE Plasma 5.21.4, Display Server: X Server 1.20.11, Display Driver: NVIDIA 465.27, OpenGL: 4.6.0, Vulkan: 1.2.168, Compiler: GCC 10.2.0 + Clang 13.0.0, File-System: btrfs, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: always
Environment Notes: __GL_ExperimentalPerfStrategy=1 __GL_IGNORE_GLSL_EXT_REQS=1 __GL_THREADED_OPTIMIZATIONS=1 __GL_SHADER_DISK_CACHE=1 __GL_SHADER_CACHE=1 __GL_ALLOW_UNOFFICIAL_PROTOCOL=1
Disk Notes: BFQ / commit=120,compress=zstd:3,discard=async,lazytime,noatime,rw,space_cache,ssd_spread,subvol=/@home,subvolid=258 / Block Size: 4096
Processor Notes: Scaling Governor: acpi-cpufreq performance - CPU Microcode: 0x10000c8
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Not affected + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable STIBP: disabled + srbds: Not affected + tsx_async_abort: Not affected

Kernel GCC

OS: ManjaroLinux 21.0.4, Kernel: 5.12.2-153-tkg-pds (x86_64), Desktop: KDE Plasma 5.21.4, Display Server: X Server 1.20.11, Display Driver: NVIDIA 465.27, OpenGL: 4.6.0, Vulkan: 1.2.168, Compiler: GCC 10.2.0 + Clang 13.0.0, File-System: btrfs, Screen Resolution: 1920x1080

Kernel LLVM 11.1

OS: ManjaroLinux 21.0.4, Kernel: 5.12.2-153-tkg-pds-llvm (x86_64), Desktop: KDE Plasma 5.21.4, Display Server: X Server 1.20.11, Display Driver: NVIDIA 465.27, OpenGL: 4.6.0, Vulkan: 1.2.168, Compiler: GCC 10.2.0 + Clang 11.1.0 + LLVM 11.1.0, File-System: btrfs, Screen Resolution: 1920x1080

Kernel GCC 11.1.1

OS: ManjaroLinux 21.0.4, Kernel: 5.12.2-154-tkg-pds (x86_64), Desktop: KDE Plasma 5.21.4, Display Server: X Server 1.20.11, Display Driver: NVIDIA 465.27, OpenGL: 4.6.0, Vulkan: 1.2.168, Compiler: GCC 10.2.0 + Clang 11.1.0 + LLVM 11.1.0, File-System: btrfs, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: always
Environment Notes: __GL_ExperimentalPerfStrategy=1 LTO_EXTRA_CFLAGS="-fgraphite-identity -floop-nest-optimize -fgcse-las -fgcse-sm -floop-interchange -floop-strip-mine -floop-block -ftree-loop-distribution -fno-signed-zeros -fno-trapping-math -fassociative-math -freciprocal-math -fno-math-errno -ffinite-math-only -fno-rounding-math -fno-signaling-nans -fcx-limited-range -fexcess-precision=fast -fdevirtualize-at-ltrans -feliminate-unused-debug-types -ftree-vectorize -mfpmath=sse -fno-semantic-interposition -ftree-loop-distribute-patterns -malign-data=cacheline -ftree-loop-vectorize -fbranch-target-load-optimize -fno-common" __GL_IGNORE_GLSL_EXT_REQS=1 __GL_THREADED_OPTIMIZATIONS=1 __GL_SHADER_DISK_CACHE=1 __GL_SHADER_CACHE=1 __GL_ALLOW_UNOFFICIAL_PROTOCOL=1
Disk Notes: BFQ / commit=120,compress=zstd:3,discard=async,lazytime,noatime,rw,space_cache,ssd_spread,subvol=/@home,subvolid=258 / Block Size: 4096
Processor Notes: Scaling Governor: acpi-cpufreq performance - CPU Microcode: 0x10000c8
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Not affected + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable STIBP: disabled + srbds: Not affected + tsx_async_abort: Not affected

Kernel LLVMProcessorMotherboardChipsetMemoryDiskGraphicsAudioNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLVulkanCompilerFile-SystemScreen ResolutionKernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.1AMD Phenom II X4 965 @ 3.30GHz (4 Cores)MSI 890FXA-GD70 (MS-7640) v1.0 (V1.8 BIOS)AMD RD890 PCI-e GFX Hydra part16GB116GB Samsung SSD 840 + 200GB Seagate ST3200822ASMSI NVIDIA GeForce GTX 650 Ti 1GBRealtek ALC892Realtek RTL8111/8168/8411ManjaroLinux 21.0.45.12.2-153-tkg-pds-llvm (x86_64)KDE Plasma 5.21.4X Server 1.20.11NVIDIA 465.274.6.01.2.168GCC 10.2.0 + Clang 13.0.0btrfs1920x10805.12.2-153-tkg-pds (x86_64)5.12.2-153-tkg-pds-llvm (x86_64)GCC 10.2.0 + Clang 11.1.0 + LLVM 11.1.05.12.2-154-tkg-pds (x86_64)OpenBenchmarking.orgKernel Details- Transparent Huge Pages: alwaysEnvironment Details- Kernel LLVM: __GL_ExperimentalPerfStrategy=1 __GL_IGNORE_GLSL_EXT_REQS=1 __GL_THREADED_OPTIMIZATIONS=1 __GL_SHADER_DISK_CACHE=1 __GL_SHADER_CACHE=1 __GL_ALLOW_UNOFFICIAL_PROTOCOL=1 - Kernel GCC: __GL_ExperimentalPerfStrategy=1 __GL_IGNORE_GLSL_EXT_REQS=1 __GL_THREADED_OPTIMIZATIONS=1 __GL_SHADER_DISK_CACHE=1 __GL_SHADER_CACHE=1 __GL_ALLOW_UNOFFICIAL_PROTOCOL=1 - Kernel LLVM 11.1: __GL_ExperimentalPerfStrategy=1 __GL_IGNORE_GLSL_EXT_REQS=1 __GL_THREADED_OPTIMIZATIONS=1 __GL_SHADER_DISK_CACHE=1 __GL_SHADER_CACHE=1 __GL_ALLOW_UNOFFICIAL_PROTOCOL=1 - Kernel GCC 11.1.1: __GL_ExperimentalPerfStrategy=1 LTO_EXTRA_CFLAGS="-fgraphite-identity -floop-nest-optimize -fgcse-las -fgcse-sm -floop-interchange -floop-strip-mine -floop-block -ftree-loop-distribution -fno-signed-zeros -fno-trapping-math -fassociative-math -freciprocal-math -fno-math-errno -ffinite-math-only -fno-rounding-math -fno-signaling-nans -fcx-limited-range -fexcess-precision=fast -fdevirtualize-at-ltrans -feliminate-unused-debug-types -ftree-vectorize -mfpmath=sse -fno-semantic-interposition -ftree-loop-distribute-patterns -malign-data=cacheline -ftree-loop-vectorize -fbranch-target-load-optimize -fno-common" __GL_IGNORE_GLSL_EXT_REQS=1 __GL_THREADED_OPTIMIZATIONS=1 __GL_SHADER_DISK_CACHE=1 __GL_SHADER_CACHE=1 __GL_ALLOW_UNOFFICIAL_PROTOCOL=1 Disk Details- BFQ / commit=120,compress=zstd:3,discard=async,lazytime,noatime,rw,space_cache,ssd_spread,subvol=/@home,subvolid=258 / Block Size: 4096Processor Details- Scaling Governor: acpi-cpufreq performance - CPU Microcode: 0x10000c8Security Details- itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Not affected + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable STIBP: disabled + srbds: Not affected + tsx_async_abort: Not affected

Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.1Result OverviewPhoronix Test Suite100%106%112%119%125%perf-benchiPerfHackbenchFacebook RocksDBApache BenchmarkWireGuard + Linux Networking Stack Stress TestSQLite SpeedtestOSBenchPostMarkt-test1IPC_benchmarkMBWLevelDBpmbenchctx_clockStress-NGOpenSSLBenchmarkMutexTinymembenchSchbench

Kernel LLVMhackbench: 32 - Processwireguard: hackbench: 16 - Threadtinymembench: Standard Memsettinymembench: Standard Memcpyhackbench: 8 - Threadmutex: Mutex Lock Unlock spinlockpmbench: 1 - 100% Readsrocksdb: Read While Writinghackbench: 16 - Processstress-ng: MMAPsqlite-speedtest: Timed Time - Size 1,000pmbench: 4 - 80% Reads 20% Writespmbench: 2 - 100% Readsstress-ng: CPU Cachembw: Memory Copy, Fixed Block Size - 4096 MiBstress-ng: Semaphoresschbench: 8 - 4hackbench: 8 - Processmbw: Memory Copy - 4096 MiBpmbench: 1 - 80% Reads 20% Writesmutex: Mutex Lock Unlock pthread_mutexmutex: Shared Mutex Lock Sharedperf-bench: Memset 1MBpostmark: Disk Transaction Performancepmbench: 2 - 50%pmbench: 2 - 100% Writesperf-bench: Memcpy 1MBpmbench: 2 - 80% Reads 20% Writespmbench: 1 - 50%pmbench: 4 - 50%pmbench: 1 - 100% Writespmbench: 4 - 100% Writespmbench: 4 - 100% Readsrocksdb: Rand Fillrocksdb: Rand Fill Syncrocksdb: Rand Readapache: Static Web Page Servingperf-bench: Futex Lock-Pistress-ng: System V Message Passingmutex: Mutex Lock Unlock spinlock_amdpgbench: 100 - 1 - Read Only - Average Latencypgbench: 100 - 1 - Read Onlyhackbench: 4 - Threadhackbench: 2 - Processhackbench: 4 - Processmutex: Mutex Lock Unlock std::mutexstress-ng: SENDFILEstress-ng: Glibc Qsort Data Sortingt-test1: 1perf-bench: Epoll Waitiperf: 5201 - 10 Seconds - TCP - 32stress-ng: CPU Stressstress-ng: Mallocstress-ng: Cryptostress-ng: MEMFDperf-bench: Futex Hashstress-ng: Memory Copyingstress-ng: Socket Activitystress-ng: Glibc C String Functionsstress-ng: Context Switchingstress-ng: Vector Mathstress-ng: Matrix Mathstress-ng: Forkingstress-ng: Atomicmbw: Memory Copy, Fixed Block Size - 1024 MiBmutex: Mutex Lock Unlock std::mutexmutex: Mutex Lock Unlock ticket_spinlockpgbench: 1 - 250 - Read Only - Average Latencypgbench: 1 - 250 - Read Onlypgbench: 1 - 1 - Read Write - Average Latencypgbench: 1 - 1 - Read Writepgbench: 1 - 100 - Read Only - Average Latencypgbench: 1 - 100 - Read Onlypgbench: 1 - 50 - Read Only - Average Latencypgbench: 1 - 50 - Read Onlypgbench: 1 - 1 - Read Only - Average Latencypgbench: 1 - 1 - Read Onlyiperf: 5201 - 10 Seconds - TCP - 64openssl: RSA 4096-bit Performanceiperf: 5201 - 10 Seconds - UDP - 1000Mbit Objective - 64iperf: 5201 - 10 Seconds - TCP - 1perf-bench: Sched Pipehackbench: 2 - Threadmutex: Semaphore Release And Acquirembw: Memory Copy - 1024 MiBosbench: Create Threadsipc-benchmark: Unnamed Unix Domain Socket - 4096ipc-benchmark: Unnamed Pipe - 4096ipc-benchmark: FIFO Named Pipe - 4096hackbench: 1 - Processipc-benchmark: TCP Socket - 4096ipc-benchmark: Unnamed Unix Domain Socket - 2048mbw: Memory Copy, Fixed Block Size - 512 MiBipc-benchmark: Unnamed Unix Domain Socket - 512iperf: 5201 - 10 Seconds - UDP - 64iperf: 5201 - 10 Seconds - UDP - 100Mbit Objective - 64iperf: 5201 - 10 Seconds - UDP - 32iperf: 5201 - 10 Seconds - UDP - 1000Mbit Objective - 32iperf: 5201 - 10 Seconds - UDP - 100Mbit Objective - 32iperf: 5201 - 10 Seconds - UDP - 1iperf: 5201 - 10 Seconds - UDP - 1000Mbit Objective - 1iperf: 5201 - 10 Seconds - UDP - 100Mbit Objective - 1hackbench: 1 - Threadipc-benchmark: FIFO Named Pipe - 2048ipc-benchmark: Unnamed Pipe - 2048t-test1: 2ipc-benchmark: TCP Socket - 256ipc-benchmark: TCP Socket - 2048mbw: Memory Copy - 512 MiBipc-benchmark: Unnamed Pipe - 512ipc-benchmark: Unnamed Unix Domain Socket - 1024rocksdb: Seq Fillipc-benchmark: Unnamed Unix Domain Socket - 256ipc-benchmark: TCP Socket - 1024ipc-benchmark: FIFO Named Pipe - 1024ipc-benchmark: Unnamed Pipe - 1024leveldb: Seq Fillleveldb: Seq Fillleveldb: Rand Deleteipc-benchmark: Unnamed Unix Domain Socket - 128ipc-benchmark: Unnamed Pipe - 128ipc-benchmark: TCP Socket - 128perf-bench: Syscall Basicipc-benchmark: TCP Socket - 512osbench: Create Filesosbench: Memory Allocationsipc-benchmark: FIFO Named Pipe - 512osbench: Launch Programsosbench: Create Processesleveldb: Rand Readipc-benchmark: FIFO Named Pipe - 256ipc-benchmark: Unnamed Pipe - 256mbw: Memory Copy, Fixed Block Size - 128 MiBleveldb: Seek Randipc-benchmark: FIFO Named Pipe - 128leveldb: Rand Fillleveldb: Rand Fillmbw: Memory Copy - 128 MiBleveldb: Hot Readleveldb: Overwriteleveldb: Overwritectx-clock: Context Switch Timeleveldb: Fill Syncleveldb: Fill Syncstress-ng: NUMAKernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.1369.548550.068168.2615702.35825.388.29442.90.0623244439159.74912.45132.1900.12400.06406.843600.3731229760.033351979.2015105.8680.117521.331.912.03361037130.11610.10655.9691030.12000.11280.11910.10380.11320.06554520574900631312343316754.6797384225731.8346.90.0501991444.09920.91040.36135.053687.4736.8231.99936540920013825.5911148830.46644.009.0171933411019.431877.82112501.83548376.449619.9811754.1519252.07167420.303575.71925.524.94.305580837.2251381.445691960.685730130.0412433118133441.7201792263721427123.06019.05152.91116.55006445795946887546563610.3875500895203003548.362115117567.1640033.62141232001.05100010011.97980748081438711.04125399799508805146.6711772787936841534770120431214538271280989128548713.95731.613.45313635612891727298500116627566206296838.822397141.110659174667895.50730440.9309072.293227845423517613623.8633.359280115714.48130.55172.9742.31014.57030.326417.67924.938317.405513.194153.6115691.55826.177.87742.50.0632239799144.75113.13125.9620.12350.06376.903722.1061277296.322859772.8425267.7120.115921.331.611.99891238660.11760.10595.9133780.12090.11350.12060.10320.11290.06494589174996401334783418155.45131554626477.3146.639.90119.35437.05034.755850.5036.7431.27439407820700828.2411170312.20642.709.071131781014.421957.06113135.64530566.539629.6111754.9419887.81167641.363724.20425.224.818304448.0204112282220993221.53419.15257.92715.74564047355947968348351110.0845550855596653719.911117121367.1640033.62143432001.05100010010.96381157081967910.93925171339441385267.5201809330964065544572123368814213621297216131691313.54732.613.16113747453040261297134316907939205408636.750389135.699351177633391.06079739.4868852.411229970323373773715.4883.393304368614.13731.25259.7142.38914.12531.226617.43525.138349.937555.201176.7005677.35809.388.56242.50.0641247556161.29112.71131.7200.12340.06456.693689.6861274218.422874781.6845240.0970.119421.331.912.01704737130.11820.10855.9541500.12180.11380.12310.10490.11450.06724490784906661328395416830.43130434244091.6846.744.30020.88841.25535.054398.5136.4432.65637555817735824.8311061908.02644.68.8870066651014.431912.64111451.53574464.009625.4211766.6419280.77168572.723705.67825.324.816653447.8198352189921743323.49819.05252.53716.01998044814447169346966811.4585625765152453693.634112032267.1640033.62050432001.05100010012.14680865881276811.14425038899481875233.3181798509926277535796120916114716451281112127965213.72532.213.46213368172722030294322716586283204467739.395617137.383382174710397.43293140.3134032.353215034122837383648.5443.412264985514.72230.05160.3922.31714.73129.926418.05724.338351.330546.302152.8015662.75810.277.02842.80.0633236996143.68312.43129.8400.13600.06386.883689.8991197368.792939372.1645241.7370.122121.332.211.56433438860.12010.10845.9276450.12180.11420.12180.10650.11530.06483716964182031292588517756.27460454546211.1546.840.06419.70137.33834.758610.1536.5632.38741399712052825.7911072394.46640.738.6870896861021.291907.72112275.43580896.529599.1811746.1019951.99168463.993682.90225.224.911544444.1118621287619904921.931195221.13414.5665804749514835664909149.5975567045587253665.782113745367.1640033.62097432001.05100010011.39081567880920610.96024675689416455204.9241824207933478490354128103114424071324519132229413.40532.912.84714111343034098290620916699690201693039.003627143.742005183570193.32021139.4765542.402230629823601073627.9683.358298691214.15931.25171.8062.39313.91831.726817.68124.838OpenBenchmarking.org

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 32 - Type: ProcessKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC80160240320400SE +/- 6.99, N = 9SE +/- 24.00, N = 9SE +/- 9.64, N = 9SE +/- 6.91, N = 8369.55351.33349.94317.411. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 32 - Type: ProcessKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC70140210280350Min: 341.54 / Avg: 369.55 / Max: 415.12Min: 288.57 / Avg: 351.33 / Max: 534.44Min: 314.26 / Avg: 349.94 / Max: 375.28Min: 295.63 / Avg: 317.4 / Max: 345.761. (CC) gcc options: -lpthread

WireGuard + Linux Networking Stack Stress Test

This is a benchmark of the WireGuard secure VPN tunnel and Linux networking stack stress test. The test runs on the local host but does require root permissions to run. The way it works is it creates three namespaces. ns0 has a loopback device. ns1 and ns2 each have wireguard devices. Those two wireguard devices send traffic through the loopback device of ns0. The end result of this is that tests wind up testing encryption and decryption at the same time -- a pretty CPU and scheduler-heavy workflow. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterWireGuard + Linux Networking Stack Stress TestKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC120240360480600SE +/- 0.65, N = 3SE +/- 0.12, N = 3SE +/- 25.08, N = 9SE +/- 1.35, N = 3555.20550.07546.30513.19
OpenBenchmarking.orgSeconds, Fewer Is BetterWireGuard + Linux Networking Stack Stress TestKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC100200300400500Min: 553.98 / Avg: 555.2 / Max: 556.22Min: 549.9 / Avg: 550.07 / Max: 550.31Min: 518.86 / Avg: 546.3 / Max: 746.92Min: 511.51 / Avg: 513.19 / Max: 515.87

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 16 - Type: ThreadKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.14080120160200SE +/- 2.17, N = 4SE +/- 1.75, N = 3SE +/- 0.69, N = 3SE +/- 1.30, N = 8176.70168.26153.61152.801. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 16 - Type: ThreadKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.1306090120150Min: 172.52 / Avg: 176.7 / Max: 182.79Min: 165.28 / Avg: 168.26 / Max: 171.33Min: 152.81 / Avg: 153.61 / Max: 154.98Min: 148.31 / Avg: 152.8 / Max: 160.61. (CC) gcc options: -lpthread

Tinymembench

This benchmark tests the system memory (RAM) performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterTinymembench 2018-05-28Standard MemsetKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM12002400360048006000SE +/- 11.53, N = 3SE +/- 3.21, N = 3SE +/- 2.89, N = 3SE +/- 15.70, N = 35662.75677.35691.55702.31. (CC) gcc options: -O2 -lm
OpenBenchmarking.orgMB/s, More Is BetterTinymembench 2018-05-28Standard MemsetKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM10002000300040005000Min: 5645.3 / Avg: 5662.67 / Max: 5684.5Min: 5671.5 / Avg: 5677.27 / Max: 5682.6Min: 5685.9 / Avg: 5691.53 / Max: 5695.5Min: 5671.1 / Avg: 5702.27 / Max: 5721.21. (CC) gcc options: -O2 -lm

OpenBenchmarking.orgMB/s, More Is BetterTinymembench 2018-05-28Standard MemcpyKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC12002400360048006000SE +/- 6.50, N = 3SE +/- 6.44, N = 3SE +/- 10.38, N = 3SE +/- 8.19, N = 35809.35810.25825.35826.11. (CC) gcc options: -O2 -lm
OpenBenchmarking.orgMB/s, More Is BetterTinymembench 2018-05-28Standard MemcpyKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC10002000300040005000Min: 5802.5 / Avg: 5809.3 / Max: 5822.3Min: 5797.9 / Avg: 5810.17 / Max: 5819.7Min: 5804.5 / Avg: 5825.27 / Max: 5835.7Min: 5809.8 / Avg: 5826.1 / Max: 5835.61. (CC) gcc options: -O2 -lm

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 8 - Type: ThreadKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.120406080100SE +/- 0.81, N = 15SE +/- 0.97, N = 3SE +/- 0.23, N = 3SE +/- 0.50, N = 1588.5688.2977.8877.031. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 8 - Type: ThreadKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.120406080100Min: 83.68 / Avg: 88.56 / Max: 93.85Min: 86.68 / Avg: 88.29 / Max: 90.04Min: 77.55 / Avg: 77.88 / Max: 78.31Min: 74.06 / Avg: 77.03 / Max: 80.881. (CC) gcc options: -lpthread

BenchmarkMutex

BenchmarkMutex is a mutex benchmark created by Malte Skarupke. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock spinlockKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC1020304050SE +/- 0.12, N = 3SE +/- 0.12, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 342.942.842.542.51. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock spinlockKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC918273645Min: 42.7 / Avg: 42.87 / Max: 43.1Min: 42.6 / Avg: 42.8 / Max: 43Min: 42.5 / Avg: 42.5 / Max: 42.5Min: 42.5 / Avg: 42.5 / Max: 42.51. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

pmbench

Pmbench is a Linux paging and virtual memory benchmark. This test profile will report the average page latency of the system. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 1 - Read-Write Ratio: 100% ReadsKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCCKernel LLVM0.01440.02880.04320.05760.072SE +/- 0.0003, N = 3SE +/- 0.0005, N = 15SE +/- 0.0006, N = 7SE +/- 0.0005, N = 150.06410.06330.06320.06231. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread
OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 1 - Read-Write Ratio: 100% ReadsKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCCKernel LLVM12345Min: 0.06 / Avg: 0.06 / Max: 0.06Min: 0.06 / Avg: 0.06 / Max: 0.07Min: 0.06 / Avg: 0.06 / Max: 0.06Min: 0.06 / Avg: 0.06 / Max: 0.061. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Read While WritingKernel GCC 11.1.1Kernel GCCKernel LLVMKernel LLVM 11.150K100K150K200K250KSE +/- 2701.00, N = 3SE +/- 772.88, N = 3SE +/- 5812.42, N = 14SE +/- 5773.77, N = 142369962397992444392475561. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Read While WritingKernel GCC 11.1.1Kernel GCCKernel LLVMKernel LLVM 11.140K80K120K160K200KMin: 231598 / Avg: 236996 / Max: 239875Min: 238255 / Avg: 239798.67 / Max: 240640Min: 236674 / Avg: 244439.29 / Max: 319760Min: 236935 / Avg: 247555.86 / Max: 3199421. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 16 - Type: ProcessKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.14080120160200SE +/- 1.83, N = 3SE +/- 0.62, N = 3SE +/- 0.22, N = 3SE +/- 0.54, N = 3161.29159.75144.75143.681. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 16 - Type: ProcessKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.1306090120150Min: 158.26 / Avg: 161.29 / Max: 164.57Min: 159 / Avg: 159.75 / Max: 160.97Min: 144.42 / Avg: 144.75 / Max: 145.18Min: 142.66 / Avg: 143.68 / Max: 144.51. (CC) gcc options: -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MMAPKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC3691215SE +/- 0.22, N = 15SE +/- 0.25, N = 12SE +/- 0.19, N = 15SE +/- 0.22, N = 1512.4312.4512.7113.131. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MMAPKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC48121620Min: 11.23 / Avg: 12.43 / Max: 13.5Min: 11.23 / Avg: 12.45 / Max: 13.33Min: 11.69 / Avg: 12.71 / Max: 13.27Min: 11.6 / Avg: 13.13 / Max: 13.71. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

SQLite Speedtest

This is a benchmark of SQLite's speedtest1 benchmark program with an increased problem size of 1,000. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000Kernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC306090120150SE +/- 0.52, N = 3SE +/- 0.40, N = 3SE +/- 0.41, N = 3SE +/- 0.26, N = 3132.19131.72129.84125.961. (CC) gcc options: -O2 -ldl -lz -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000Kernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC20406080100Min: 131.17 / Avg: 132.19 / Max: 132.88Min: 131.24 / Avg: 131.72 / Max: 132.52Min: 129.4 / Avg: 129.84 / Max: 130.67Min: 125.65 / Avg: 125.96 / Max: 126.481. (CC) gcc options: -O2 -ldl -lz -lpthread

pmbench

Pmbench is a Linux paging and virtual memory benchmark. This test profile will report the average page latency of the system. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 4 - Read-Write Ratio: 80% Reads 20% WritesKernel GCC 11.1.1Kernel LLVMKernel GCCKernel LLVM 11.10.03060.06120.09180.12240.153SE +/- 0.0072, N = 15SE +/- 0.0016, N = 3SE +/- 0.0012, N = 3SE +/- 0.0014, N = 30.13600.12400.12350.12341. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread
OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 4 - Read-Write Ratio: 80% Reads 20% WritesKernel GCC 11.1.1Kernel LLVMKernel GCCKernel LLVM 11.112345Min: 0.12 / Avg: 0.14 / Max: 0.23Min: 0.12 / Avg: 0.12 / Max: 0.13Min: 0.12 / Avg: 0.12 / Max: 0.12Min: 0.12 / Avg: 0.12 / Max: 0.131. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 2 - Read-Write Ratio: 100% ReadsKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC0.01450.0290.04350.0580.0725SE +/- 0.0008, N = 3SE +/- 0.0000, N = 3SE +/- 0.0006, N = 15SE +/- 0.0000, N = 30.06450.06400.06380.06371. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread
OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 2 - Read-Write Ratio: 100% ReadsKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC12345Min: 0.06 / Avg: 0.06 / Max: 0.07Min: 0.06 / Avg: 0.06 / Max: 0.06Min: 0.06 / Avg: 0.06 / Max: 0.07Min: 0.06 / Avg: 0.06 / Max: 0.061. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CPU CacheKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC246810SE +/- 0.11, N = 15SE +/- 0.06, N = 15SE +/- 0.03, N = 3SE +/- 0.12, N = 156.696.846.886.901. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CPU CacheKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC3691215Min: 6.17 / Avg: 6.69 / Max: 7.6Min: 6.27 / Avg: 6.84 / Max: 7.13Min: 6.83 / Avg: 6.88 / Max: 6.93Min: 6 / Avg: 6.9 / Max: 7.471. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiBKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC8001600240032004000SE +/- 37.05, N = 3SE +/- 3.81, N = 3SE +/- 2.63, N = 3SE +/- 0.67, N = 33600.373689.693689.903722.111. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 4096 MiBKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC6001200180024003000Min: 3557.63 / Avg: 3600.37 / Max: 3674.18Min: 3685.04 / Avg: 3689.69 / Max: 3697.24Min: 3685.57 / Avg: 3689.9 / Max: 3694.64Min: 3720.83 / Avg: 3722.11 / Max: 3723.081. (CC) gcc options: -O3 -march=native

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: SemaphoresKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC300K600K900K1200K1500KSE +/- 13131.30, N = 3SE +/- 10868.11, N = 15SE +/- 14472.05, N = 3SE +/- 9507.55, N = 151197368.791229760.031274218.421277296.321. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: SemaphoresKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC200K400K600K800K1000KMin: 1173226.95 / Avg: 1197368.79 / Max: 1218393.56Min: 1156406.91 / Avg: 1229760.03 / Max: 1297438.51Min: 1253601.69 / Avg: 1274218.42 / Max: 1302120.34Min: 1237913.92 / Avg: 1277296.32 / Max: 1346381.211. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

Schbench

This is a benchmark of Schbench, a Linux kernel scheduler benchmark developed by Facebook. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgusec, 99.9th Latency Percentile, Fewer Is BetterSchbenchMessage Threads: 8 - Workers Per Message Thread: 4Kernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC7K14K21K28K35KSE +/- 2786.67, N = 15SE +/- 386.40, N = 13SE +/- 216.51, N = 3SE +/- 267.31, N = 3335192939328747285971. (CC) gcc options: -O2 -lpthread
OpenBenchmarking.orgusec, 99.9th Latency Percentile, Fewer Is BetterSchbenchMessage Threads: 8 - Workers Per Message Thread: 4Kernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC6K12K18K24K30KMin: 26016 / Avg: 33518.93 / Max: 58432Min: 27616 / Avg: 29393.23 / Max: 32096Min: 28320 / Avg: 28746.67 / Max: 29024Min: 28064 / Avg: 28597.33 / Max: 288961. (CC) gcc options: -O2 -lpthread

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 8 - Type: ProcessKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.120406080100SE +/- 0.85, N = 4SE +/- 0.46, N = 3SE +/- 0.85, N = 3SE +/- 0.33, N = 381.6879.2072.8472.161. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 8 - Type: ProcessKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.11632486480Min: 79.31 / Avg: 81.68 / Max: 83.37Min: 78.3 / Avg: 79.2 / Max: 79.76Min: 71.49 / Avg: 72.84 / Max: 74.4Min: 71.59 / Avg: 72.16 / Max: 72.721. (CC) gcc options: -lpthread

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 4096 MiBKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC11002200330044005500SE +/- 62.83, N = 3SE +/- 9.07, N = 3SE +/- 4.84, N = 3SE +/- 5.16, N = 35105.875240.105241.745267.711. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 4096 MiBKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC9001800270036004500Min: 5000.14 / Avg: 5105.87 / Max: 5217.54Min: 5224.93 / Avg: 5240.1 / Max: 5256.29Min: 5232.09 / Avg: 5241.74 / Max: 5247.26Min: 5260.99 / Avg: 5267.71 / Max: 5277.861. (CC) gcc options: -O3 -march=native

pmbench

Pmbench is a Linux paging and virtual memory benchmark. This test profile will report the average page latency of the system. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 1 - Read-Write Ratio: 80% Reads 20% WritesKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC0.02750.0550.08250.110.1375SE +/- 0.0012, N = 5SE +/- 0.0010, N = 3SE +/- 0.0012, N = 4SE +/- 0.0011, N = 30.12210.11940.11750.11591. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread
OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 1 - Read-Write Ratio: 80% Reads 20% WritesKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC12345Min: 0.12 / Avg: 0.12 / Max: 0.13Min: 0.12 / Avg: 0.12 / Max: 0.12Min: 0.11 / Avg: 0.12 / Max: 0.12Min: 0.11 / Avg: 0.12 / Max: 0.121. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

BenchmarkMutex

BenchmarkMutex is a mutex benchmark created by Malte Skarupke. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock pthread_mutexKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM510152025SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.07, N = 3SE +/- 0.03, N = 321.321.321.321.31. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock pthread_mutexKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM510152025Min: 21.3 / Avg: 21.3 / Max: 21.3Min: 21.3 / Avg: 21.3 / Max: 21.3Min: 21.2 / Avg: 21.33 / Max: 21.4Min: 21.3 / Avg: 21.33 / Max: 21.41. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Shared Mutex Lock SharedKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC714212835SE +/- 0.27, N = 14SE +/- 0.15, N = 3SE +/- 0.09, N = 3SE +/- 0.07, N = 332.231.931.931.61. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Shared Mutex Lock SharedKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC714212835Min: 31.5 / Avg: 32.22 / Max: 35.4Min: 31.7 / Avg: 31.93 / Max: 32.2Min: 31.7 / Avg: 31.87 / Max: 32Min: 31.5 / Avg: 31.57 / Max: 31.71. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memset 1MBKernel GCC 11.1.1Kernel GCCKernel LLVM 11.1Kernel LLVM3691215SE +/- 0.11, N = 15SE +/- 0.05, N = 3SE +/- 0.04, N = 3SE +/- 0.01, N = 311.5612.0012.0212.031. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma
OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memset 1MBKernel GCC 11.1.1Kernel GCCKernel LLVM 11.1Kernel LLVM48121620Min: 10.7 / Avg: 11.56 / Max: 12.09Min: 11.95 / Avg: 12 / Max: 12.09Min: 11.97 / Avg: 12.02 / Max: 12.1Min: 12.02 / Avg: 12.03 / Max: 12.041. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma

PostMark

This is a test of NetApp's PostMark benchmark designed to simulate small-file testing similar to the tasks endured by web and mail servers. This test profile will set PostMark to perform 25,000 transactions with 500 files simultaneously with the file sizes ranging between 5 and 512 kilobytes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostMark 1.51Disk Transaction PerformanceKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.18001600240032004000SE +/- 18.33, N = 3SE +/- 18.33, N = 3SE +/- 20.00, N = 3SE +/- 20.00, N = 337133713386638861. (CC) gcc options: -O3
OpenBenchmarking.orgTPS, More Is BetterPostMark 1.51Disk Transaction PerformanceKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.17001400210028003500Min: 3676 / Avg: 3712.67 / Max: 3731Min: 3676 / Avg: 3712.67 / Max: 3731Min: 3846 / Avg: 3866 / Max: 3906Min: 3846 / Avg: 3886 / Max: 39061. (CC) gcc options: -O3

pmbench

Pmbench is a Linux paging and virtual memory benchmark. This test profile will report the average page latency of the system. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 2 - Read-Write Ratio: 50%Kernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM0.0270.0540.0810.1080.135SE +/- 0.0015, N = 3SE +/- 0.0000, N = 3SE +/- 0.0014, N = 4SE +/- 0.0014, N = 30.12010.11820.11760.11611. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread
OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 2 - Read-Write Ratio: 50%Kernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM12345Min: 0.12 / Avg: 0.12 / Max: 0.12Min: 0.12 / Avg: 0.12 / Max: 0.12Min: 0.11 / Avg: 0.12 / Max: 0.12Min: 0.11 / Avg: 0.12 / Max: 0.121. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 2 - Read-Write Ratio: 100% WritesKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC0.02440.04880.07320.09760.122SE +/- 0.0009, N = 3SE +/- 0.0013, N = 4SE +/- 0.0014, N = 3SE +/- 0.0011, N = 30.10850.10840.10650.10591. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread
OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 2 - Read-Write Ratio: 100% WritesKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC12345Min: 0.11 / Avg: 0.11 / Max: 0.11Min: 0.1 / Avg: 0.11 / Max: 0.11Min: 0.1 / Avg: 0.11 / Max: 0.11Min: 0.1 / Avg: 0.11 / Max: 0.111. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memcpy 1MBKernel GCCKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVM1.3432.6864.0295.3726.715SE +/- 0.051475, N = 3SE +/- 0.072648, N = 3SE +/- 0.003067, N = 3SE +/- 0.007854, N = 35.9133785.9276455.9541505.9691031. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma
OpenBenchmarking.orgGB/sec, More Is Betterperf-benchBenchmark: Memcpy 1MBKernel GCCKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVM246810Min: 5.81 / Avg: 5.91 / Max: 5.99Min: 5.78 / Avg: 5.93 / Max: 6.02Min: 5.95 / Avg: 5.95 / Max: 5.96Min: 5.96 / Avg: 5.97 / Max: 5.981. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma

pmbench

Pmbench is a Linux paging and virtual memory benchmark. This test profile will report the average page latency of the system. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 2 - Read-Write Ratio: 80% Reads 20% WritesKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM0.02740.05480.08220.10960.137SE +/- 0.0011, N = 3SE +/- 0.0004, N = 3SE +/- 0.0003, N = 3SE +/- 0.0012, N = 30.12180.12180.12090.12001. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread
OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 2 - Read-Write Ratio: 80% Reads 20% WritesKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM12345Min: 0.12 / Avg: 0.12 / Max: 0.12Min: 0.12 / Avg: 0.12 / Max: 0.12Min: 0.12 / Avg: 0.12 / Max: 0.12Min: 0.12 / Avg: 0.12 / Max: 0.121. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 1 - Read-Write Ratio: 50%Kernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM0.02570.05140.07710.10280.1285SE +/- 0.0008, N = 3SE +/- 0.0007, N = 3SE +/- 0.0015, N = 3SE +/- 0.0015, N = 30.11420.11380.11350.11281. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread
OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 1 - Read-Write Ratio: 50%Kernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM12345Min: 0.11 / Avg: 0.11 / Max: 0.12Min: 0.11 / Avg: 0.11 / Max: 0.12Min: 0.11 / Avg: 0.11 / Max: 0.12Min: 0.11 / Avg: 0.11 / Max: 0.121. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 4 - Read-Write Ratio: 50%Kernel LLVM 11.1Kernel GCC 11.1.1Kernel GCCKernel LLVM0.02770.05540.08310.11080.1385SE +/- 0.0016, N = 3SE +/- 0.0013, N = 3SE +/- 0.0006, N = 3SE +/- 0.0004, N = 30.12310.12180.12060.11911. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread
OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 4 - Read-Write Ratio: 50%Kernel LLVM 11.1Kernel GCC 11.1.1Kernel GCCKernel LLVM12345Min: 0.12 / Avg: 0.12 / Max: 0.13Min: 0.12 / Avg: 0.12 / Max: 0.12Min: 0.12 / Avg: 0.12 / Max: 0.12Min: 0.12 / Avg: 0.12 / Max: 0.121. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 1 - Read-Write Ratio: 100% WritesKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC0.0240.0480.0720.0960.12SE +/- 0.0012, N = 3SE +/- 0.0001, N = 3SE +/- 0.0012, N = 3SE +/- 0.0010, N = 30.10650.10490.10380.10321. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread
OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 1 - Read-Write Ratio: 100% WritesKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC12345Min: 0.11 / Avg: 0.11 / Max: 0.11Min: 0.1 / Avg: 0.1 / Max: 0.11Min: 0.1 / Avg: 0.1 / Max: 0.11Min: 0.1 / Avg: 0.1 / Max: 0.11. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 4 - Read-Write Ratio: 100% WritesKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC0.02590.05180.07770.10360.1295SE +/- 0.0005, N = 3SE +/- 0.0013, N = 3SE +/- 0.0014, N = 3SE +/- 0.0001, N = 30.11530.11450.11320.11291. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread
OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 4 - Read-Write Ratio: 100% WritesKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC12345Min: 0.11 / Avg: 0.12 / Max: 0.12Min: 0.11 / Avg: 0.11 / Max: 0.12Min: 0.11 / Avg: 0.11 / Max: 0.12Min: 0.11 / Avg: 0.11 / Max: 0.111. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 4 - Read-Write Ratio: 100% ReadsKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.10.01510.03020.04530.06040.0755SE +/- 0.0006, N = 3SE +/- 0.0007, N = 3SE +/- 0.0008, N = 3SE +/- 0.0006, N = 30.06720.06550.06490.06481. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread
OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 4 - Read-Write Ratio: 100% ReadsKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.112345Min: 0.07 / Avg: 0.07 / Max: 0.07Min: 0.06 / Avg: 0.07 / Max: 0.07Min: 0.06 / Avg: 0.06 / Max: 0.07Min: 0.06 / Avg: 0.06 / Max: 0.071. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random FillKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC100K200K300K400K500KSE +/- 4304.32, N = 3SE +/- 3376.28, N = 3SE +/- 3333.72, N = 3SE +/- 3750.21, N = 33716964490784520574589171. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random FillKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC80K160K240K320K400KMin: 363164 / Avg: 371696.33 / Max: 376953Min: 442408 / Avg: 449077.67 / Max: 453326Min: 448691 / Avg: 452056.67 / Max: 458724Min: 451580 / Avg: 458917 / Max: 4639341. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random Fill SyncKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC110K220K330K440K550KSE +/- 5443.53, N = 3SE +/- 2169.84, N = 3SE +/- 4275.49, N = 3SE +/- 1981.98, N = 34182034900634906664996401. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random Fill SyncKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC90K180K270K360K450KMin: 407327 / Avg: 418203.33 / Max: 424060Min: 487332 / Avg: 490062.67 / Max: 494349Min: 482738 / Avg: 490665.67 / Max: 497405Min: 495781 / Avg: 499639.67 / Max: 5023551. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random ReadKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC3M6M9M12M15MSE +/- 32869.63, N = 3SE +/- 59134.09, N = 3SE +/- 47642.98, N = 3SE +/- 85590.38, N = 3129258851312343313283954133478341. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random ReadKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC2M4M6M8M10MMin: 12888446 / Avg: 12925885.33 / Max: 12991402Min: 13039780 / Avg: 13123432.67 / Max: 13237662Min: 13214481 / Avg: 13283954.33 / Max: 13375168Min: 13177621 / Avg: 13347833.67 / Max: 134486841. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

Apache Benchmark

This is a test of ab, which is the Apache benchmark program. This test profile measures how many requests per second a given system can sustain when carrying out 1,000,000 requests with 100 requests being carried out concurrently. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.29Static Web Page ServingKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC4K8K12K16K20KSE +/- 28.17, N = 3SE +/- 18.45, N = 3SE +/- 221.42, N = 3SE +/- 31.87, N = 316754.6716830.4317756.2718155.451. (CC) gcc options: -shared -fPIC -O2 -pthread
OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.29Static Web Page ServingKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC3K6K9K12K15KMin: 16698.34 / Avg: 16754.67 / Max: 16783.87Min: 16808.99 / Avg: 16830.43 / Max: 16867.17Min: 17313.73 / Avg: 17756.27 / Max: 17991.67Min: 18115.03 / Avg: 18155.45 / Max: 18218.331. (CC) gcc options: -shared -fPIC -O2 -pthread

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex Lock-PiKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.110K20K30K40K50KSE +/- 7.69, N = 3SE +/- 12.14, N = 3SE +/- 20.00, N = 3SE +/- 2731.13, N = 1297381304313155460451. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex Lock-PiKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.18K16K24K32K40KMin: 9727 / Avg: 9738.33 / Max: 9753Min: 13021 / Avg: 13042.67 / Max: 13063Min: 13122 / Avg: 13154.67 / Max: 13191Min: 18892 / Avg: 46045 / Max: 506241. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: System V Message PassingKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC1000K2000K3000K4000K5000KSE +/- 2913.62, N = 3SE +/- 18462.77, N = 3SE +/- 19986.42, N = 3SE +/- 32996.49, N = 124225731.834244091.684546211.154626477.311. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: System V Message PassingKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC800K1600K2400K3200K4000KMin: 4222551.12 / Avg: 4225731.83 / Max: 4231550.64Min: 4207179.25 / Avg: 4244091.68 / Max: 4263400.08Min: 4515445.31 / Avg: 4546211.15 / Max: 4583695.39Min: 4365663.46 / Avg: 4626477.31 / Max: 4728468.331. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

BenchmarkMutex

BenchmarkMutex is a mutex benchmark created by Malte Skarupke. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock spinlock_amdKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC1122334455SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.00, N = 346.946.846.746.61. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock spinlock_amdKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC1020304050Min: 46.8 / Avg: 46.93 / Max: 47Min: 46.7 / Avg: 46.83 / Max: 46.9Min: 46.6 / Avg: 46.73 / Max: 46.8Min: 46.6 / Avg: 46.6 / Max: 46.61. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read Only - Average LatencyKernel LLVM0.01130.02260.03390.04520.0565SE +/- 0.000, N = 30.0501. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 100 - Clients: 1 - Mode: Read OnlyKernel LLVM4K8K12K16K20KSE +/- 112.18, N = 3199141. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 4 - Type: ThreadKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC1020304050SE +/- 0.11, N = 3SE +/- 0.16, N = 3SE +/- 0.04, N = 3SE +/- 0.35, N = 344.3044.1040.0639.901. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 4 - Type: ThreadKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC918273645Min: 44.09 / Avg: 44.3 / Max: 44.44Min: 43.84 / Avg: 44.1 / Max: 44.4Min: 40.02 / Avg: 40.06 / Max: 40.14Min: 39.43 / Avg: 39.9 / Max: 40.591. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 2 - Type: ProcessKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC510152025SE +/- 0.22, N = 14SE +/- 0.04, N = 3SE +/- 0.26, N = 3SE +/- 0.03, N = 320.9120.8919.7019.351. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 2 - Type: ProcessKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC510152025Min: 20.53 / Avg: 20.91 / Max: 23.8Min: 20.82 / Avg: 20.89 / Max: 20.93Min: 19.3 / Avg: 19.7 / Max: 20.19Min: 19.31 / Avg: 19.35 / Max: 19.41. (CC) gcc options: -lpthread

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 4 - Type: ProcessKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC918273645SE +/- 0.16, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 3SE +/- 0.12, N = 341.2640.3637.3437.051. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 4 - Type: ProcessKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC918273645Min: 40.95 / Avg: 41.26 / Max: 41.5Min: 40.27 / Avg: 40.36 / Max: 40.52Min: 37.24 / Avg: 37.34 / Max: 37.41Min: 36.82 / Avg: 37.05 / Max: 37.241. (CC) gcc options: -lpthread

BenchmarkMutex

BenchmarkMutex is a mutex benchmark created by Malte Skarupke. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock std::mutexKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC816243240SE +/- 0.13, N = 3SE +/- 0.12, N = 3SE +/- 0.07, N = 3SE +/- 0.12, N = 335.035.034.734.71. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock std::mutexKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC714212835Min: 34.9 / Avg: 35.03 / Max: 35.3Min: 34.8 / Avg: 35.03 / Max: 35.2Min: 34.6 / Avg: 34.73 / Max: 34.8Min: 34.5 / Avg: 34.73 / Max: 34.91. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: SENDFILEKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.113K26K39K52K65KSE +/- 107.63, N = 3SE +/- 46.31, N = 3SE +/- 144.65, N = 3SE +/- 600.66, N = 553687.4754398.5155850.5058610.151. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: SENDFILEKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.110K20K30K40K50KMin: 53472.23 / Avg: 53687.47 / Max: 53798.35Min: 54342.3 / Avg: 54398.51 / Max: 54490.36Min: 55673.86 / Avg: 55850.5 / Max: 56137.24Min: 56208.65 / Avg: 58610.15 / Max: 59258.661. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Glibc Qsort Data SortingKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCCKernel LLVM816243240SE +/- 0.44, N = 4SE +/- 0.47, N = 3SE +/- 0.37, N = 3SE +/- 0.24, N = 336.4436.5636.7436.821. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Glibc Qsort Data SortingKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCCKernel LLVM816243240Min: 35.7 / Avg: 36.44 / Max: 37.7Min: 35.83 / Avg: 36.56 / Max: 37.43Min: 36.23 / Avg: 36.74 / Max: 37.47Min: 36.5 / Avg: 36.82 / Max: 37.31. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

t-test1

This is a test of t-test1 for basic memory allocator benchmarks. Note this test profile is currently very basic and the overall time does include the warmup time of the custom t-test1 compilation. Improvements welcome. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Bettert-test1 2017-01-13Threads: 1Kernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC816243240SE +/- 0.05, N = 3SE +/- 0.29, N = 3SE +/- 0.02, N = 3SE +/- 0.04, N = 332.6632.3932.0031.271. (CC) gcc options: -pthread
OpenBenchmarking.orgSeconds, Fewer Is Bettert-test1 2017-01-13Threads: 1Kernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC714212835Min: 32.56 / Avg: 32.66 / Max: 32.73Min: 32 / Avg: 32.39 / Max: 32.96Min: 31.96 / Avg: 32 / Max: 32.05Min: 31.21 / Avg: 31.27 / Max: 31.341. (CC) gcc options: -pthread

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Epoll WaitKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.190K180K270K360K450KSE +/- 2596.83, N = 3SE +/- 1651.40, N = 3SE +/- 4664.03, N = 3SE +/- 1873.36, N = 33654093755583940784139971. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Epoll WaitKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.170K140K210K280K350KMin: 362252 / Avg: 365409 / Max: 370559Min: 372562 / Avg: 375557.67 / Max: 378260Min: 385422 / Avg: 394077.67 / Max: 401417Min: 410574 / Avg: 413996.67 / Max: 4170281. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma

iPerf

iPerf is a network bandwidth throughput testing software. This test profile is used for automated testing of an iperf client and requires you have access to an iperf server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: TCP - Parallel: 32Kernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC4K8K12K16K20KSE +/- 142.34, N = 3SE +/- 647.29, N = 15SE +/- 187.37, N = 7SE +/- 135.39, N = 3120521773520013207001. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: TCP - Parallel: 32Kernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC4K8K12K16K20KMin: 11908 / Avg: 12052.33 / Max: 12337Min: 13665 / Avg: 17735.2 / Max: 20235Min: 18948 / Avg: 20013.14 / Max: 20367Min: 20429 / Avg: 20699.67 / Max: 208421. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CPU StressKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC2004006008001000SE +/- 1.66, N = 3SE +/- 2.22, N = 3SE +/- 0.17, N = 3SE +/- 0.36, N = 3824.83825.59825.79828.241. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CPU StressKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC150300450600750Min: 821.58 / Avg: 824.83 / Max: 827.04Min: 821.18 / Avg: 825.59 / Max: 828.18Min: 825.49 / Avg: 825.79 / Max: 826.09Min: 827.54 / Avg: 828.24 / Max: 828.71. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MallocKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC2M4M6M8M10MSE +/- 10798.66, N = 3SE +/- 22435.78, N = 3SE +/- 12093.70, N = 3SE +/- 13169.62, N = 311061908.0211072394.4611148830.4611170312.201. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MallocKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC2M4M6M8M10MMin: 11047793.5 / Avg: 11061908.02 / Max: 11083122.21Min: 11042825.46 / Avg: 11072394.46 / Max: 11116408.2Min: 11133567.12 / Avg: 11148830.46 / Max: 11172711.6Min: 11155246.4 / Avg: 11170312.2 / Max: 11196555.591. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CryptoKernel GCC 11.1.1Kernel GCCKernel LLVMKernel LLVM 11.1140280420560700SE +/- 2.63, N = 3SE +/- 0.12, N = 3SE +/- 1.45, N = 3SE +/- 1.44, N = 3640.73642.70644.00644.601. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CryptoKernel GCC 11.1.1Kernel GCCKernel LLVMKernel LLVM 11.1110220330440550Min: 635.78 / Avg: 640.73 / Max: 644.75Min: 642.49 / Avg: 642.7 / Max: 642.92Min: 641.39 / Avg: 644 / Max: 646.4Min: 641.8 / Avg: 644.6 / Max: 646.61. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MEMFDKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM3691215SE +/- 0.02, N = 3SE +/- 0.01, N = 3SE +/- 0.03, N = 3SE +/- 0.02, N = 38.688.889.009.011. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: MEMFDKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM3691215Min: 8.63 / Avg: 8.68 / Max: 8.7Min: 8.87 / Avg: 8.88 / Max: 8.9Min: 8.9 / Avg: 8.97 / Max: 9Min: 8.96 / Avg: 9.01 / Max: 9.031. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex HashKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCCKernel LLVM1.5M3M4.5M6M7.5MSE +/- 9731.70, N = 3SE +/- 11648.66, N = 3SE +/- 5828.57, N = 3SE +/- 9923.07, N = 370066657089686711317871933411. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Futex HashKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCCKernel LLVM1.2M2.4M3.6M4.8M6MMin: 6990489 / Avg: 7006665.33 / Max: 7024127Min: 7066555 / Avg: 7089686 / Max: 7103658Min: 7106448 / Avg: 7113178 / Max: 7124786Min: 7173512 / Avg: 7193340.67 / Max: 72039761. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Memory CopyingKernel GCCKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.12004006008001000SE +/- 1.52, N = 3SE +/- 2.74, N = 3SE +/- 6.09, N = 3SE +/- 2.92, N = 31014.421014.431019.431021.291. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Memory CopyingKernel GCCKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.12004006008001000Min: 1012.57 / Avg: 1014.42 / Max: 1017.43Min: 1010.72 / Avg: 1014.43 / Max: 1019.78Min: 1012.19 / Avg: 1019.43 / Max: 1031.53Min: 1015.46 / Avg: 1021.29 / Max: 1024.421. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Socket ActivityKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC400800120016002000SE +/- 1.75, N = 3SE +/- 3.56, N = 3SE +/- 1.60, N = 3SE +/- 2.69, N = 31877.821907.721912.641957.061. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Socket ActivityKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC30060090012001500Min: 1874.64 / Avg: 1877.82 / Max: 1880.68Min: 1901.76 / Avg: 1907.72 / Max: 1914.06Min: 1909.56 / Avg: 1912.64 / Max: 1914.92Min: 1953.95 / Avg: 1957.06 / Max: 1962.431. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Glibc C String FunctionsKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC20K40K60K80K100KSE +/- 415.80, N = 3SE +/- 70.50, N = 3SE +/- 495.62, N = 3SE +/- 236.51, N = 3111451.53112275.43112501.83113135.641. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Glibc C String FunctionsKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC20K40K60K80K100KMin: 110909.37 / Avg: 111451.53 / Max: 112268.7Min: 112146.99 / Avg: 112275.43 / Max: 112390.05Min: 111636.02 / Avg: 112501.83 / Max: 113352.72Min: 112677.46 / Avg: 113135.64 / Max: 113466.541. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Context SwitchingKernel GCCKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1120K240K360K480K600KSE +/- 4193.19, N = 3SE +/- 2752.66, N = 3SE +/- 2034.95, N = 3SE +/- 5035.55, N = 3530566.53548376.44574464.00580896.521. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Context SwitchingKernel GCCKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1100K200K300K400K500KMin: 522389.48 / Avg: 530566.53 / Max: 536267.59Min: 543977.45 / Avg: 548376.44 / Max: 553442.65Min: 571999.64 / Avg: 574464 / Max: 578501.22Min: 572311.55 / Avg: 580896.52 / Max: 589749.061. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Vector MathKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC2K4K6K8K10KSE +/- 5.68, N = 3SE +/- 10.59, N = 3SE +/- 0.92, N = 3SE +/- 1.82, N = 39599.189619.989625.429629.611. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Vector MathKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC2K4K6K8K10KMin: 9587.99 / Avg: 9599.18 / Max: 9606.52Min: 9598.81 / Avg: 9619.98 / Max: 9631.36Min: 9623.77 / Avg: 9625.42 / Max: 9626.95Min: 9626.15 / Avg: 9629.61 / Max: 9632.351. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Matrix MathKernel GCC 11.1.1Kernel LLVMKernel GCCKernel LLVM 11.13K6K9K12K15KSE +/- 4.70, N = 3SE +/- 16.83, N = 3SE +/- 17.54, N = 3SE +/- 0.54, N = 311746.1011754.1511754.9411766.641. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Matrix MathKernel GCC 11.1.1Kernel LLVMKernel GCCKernel LLVM 11.12K4K6K8K10KMin: 11737.74 / Avg: 11746.1 / Max: 11754.01Min: 11720.94 / Avg: 11754.15 / Max: 11775.48Min: 11719.94 / Avg: 11754.94 / Max: 11774.54Min: 11765.68 / Avg: 11766.64 / Max: 11767.551. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: ForkingKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.14K8K12K16K20KSE +/- 39.67, N = 3SE +/- 97.53, N = 3SE +/- 41.17, N = 3SE +/- 62.92, N = 319252.0719280.7719887.8119951.991. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: ForkingKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.13K6K9K12K15KMin: 19182.55 / Avg: 19252.07 / Max: 19319.95Min: 19145.31 / Avg: 19280.77 / Max: 19470.05Min: 19842.29 / Avg: 19887.81 / Max: 19969.99Min: 19851.14 / Avg: 19951.99 / Max: 20067.61. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: AtomicKernel LLVMKernel GCCKernel GCC 11.1.1Kernel LLVM 11.140K80K120K160K200KSE +/- 324.01, N = 3SE +/- 368.77, N = 3SE +/- 187.43, N = 3SE +/- 560.17, N = 3167420.30167641.36168463.99168572.721. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: AtomicKernel LLVMKernel GCCKernel GCC 11.1.1Kernel LLVM 11.130K60K90K120K150KMin: 166907.14 / Avg: 167420.3 / Max: 168019.59Min: 167198.2 / Avg: 167641.36 / Max: 168373.51Min: 168107.38 / Avg: 168463.99 / Max: 168742.33Min: 167749.08 / Avg: 168572.72 / Max: 169642.261. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiBKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC8001600240032004000SE +/- 5.31, N = 3SE +/- 5.81, N = 3SE +/- 3.36, N = 3SE +/- 1.03, N = 33575.723682.903705.683724.201. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 1024 MiBKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC6001200180024003000Min: 3566.72 / Avg: 3575.72 / Max: 3585.1Min: 3673.18 / Avg: 3682.9 / Max: 3693.27Min: 3700 / Avg: 3705.68 / Max: 3711.62Min: 3722.73 / Avg: 3724.2 / Max: 3726.181. (CC) gcc options: -O3 -march=native

BenchmarkMutex

BenchmarkMutex is a mutex benchmark created by Malte Skarupke. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock std::mutexKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC612182430SE +/- 0.03, N = 3SE +/- 0.12, N = 3SE +/- 0.09, N = 3SE +/- 0.07, N = 325.525.325.225.21. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock std::mutexKernel LLVMKernel LLVM 11.1Kernel GCC 11.1.1Kernel GCC612182430Min: 25.4 / Avg: 25.47 / Max: 25.5Min: 25.1 / Avg: 25.3 / Max: 25.5Min: 25 / Avg: 25.17 / Max: 25.3Min: 25.1 / Avg: 25.23 / Max: 25.31. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock ticket_spinlockKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC612182430SE +/- 0.07, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 324.924.924.824.81. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Mutex Lock Unlock ticket_spinlockKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC612182430Min: 24.8 / Avg: 24.87 / Max: 25Min: 24.9 / Avg: 24.9 / Max: 24.9Min: 24.8 / Avg: 24.8 / Max: 24.8Min: 24.8 / Avg: 24.8 / Max: 24.81. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

PostgreSQL pgbench

This is a benchmark of PostgreSQL using pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 250 - Mode: Read Only - Average LatencyKernel LLVM0.96861.93722.90583.87444.843SE +/- 0.028, N = 34.3051. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 250 - Mode: Read OnlyKernel LLVM12K24K36K48K60KSE +/- 383.60, N = 3580831. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read Write - Average LatencyKernel LLVM246810SE +/- 0.086, N = 37.2251. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read WriteKernel LLVM306090120150SE +/- 1.65, N = 31381. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 100 - Mode: Read Only - Average LatencyKernel LLVM0.32510.65020.97531.30041.6255SE +/- 0.001, N = 31.4451. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 100 - Mode: Read OnlyKernel LLVM15K30K45K60K75KSE +/- 50.59, N = 3691961. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read Only - Average LatencyKernel LLVM0.15410.30820.46230.61640.7705SE +/- 0.003, N = 30.6851. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 50 - Mode: Read OnlyKernel LLVM16K32K48K64K80KSE +/- 358.11, N = 3730131. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read Only - Average LatencyKernel LLVM0.00920.01840.02760.03680.046SE +/- 0.000, N = 30.0411. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 13.0Scaling Factor: 1 - Clients: 1 - Mode: Read OnlyKernel LLVM5K10K15K20K25KSE +/- 38.99, N = 3243311. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -ldl -lm

iPerf

iPerf is a network bandwidth throughput testing software. This test profile is used for automated testing of an iperf client and requires you have access to an iperf server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: TCP - Parallel: 64Kernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC4K8K12K16K20KSE +/- 152.42, N = 3SE +/- 425.25, N = 15SE +/- 203.43, N = 4SE +/- 78.33, N = 3115441665318133183041. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: TCP - Parallel: 64Kernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC3K6K9K12K15KMin: 11357 / Avg: 11544 / Max: 11846Min: 12999 / Avg: 16653.33 / Max: 17675Min: 17689 / Avg: 18133 / Max: 18675Min: 18155 / Avg: 18304.33 / Max: 184201. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenSSL

OpenSSL is an open-source toolkit that implements SSL (Secure Sockets Layer) and TLS (Transport Layer Security) protocols. This test measures the RSA 4096-bit performance of OpenSSL. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.1.1RSA 4096-bit PerformanceKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC100200300400500SE +/- 4.10, N = 7SE +/- 2.02, N = 3SE +/- 0.03, N = 3SE +/- 0.18, N = 3441.7444.1447.8448.0
OpenBenchmarking.orgSigns Per Second, More Is BetterOpenSSL 1.1.1RSA 4096-bit PerformanceKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC80160240320400Min: 422.5 / Avg: 441.73 / Max: 448.2Min: 440.1 / Avg: 444.07 / Max: 446.7Min: 447.8 / Avg: 447.83 / Max: 447.9Min: 447.7 / Avg: 448.03 / Max: 448.3

iPerf

iPerf is a network bandwidth throughput testing software. This test profile is used for automated testing of an iperf client and requires you have access to an iperf server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - 1000Mbit Objective - Parallel: 64Kernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC4K8K12K16K20KSE +/- 948.81, N = 15SE +/- 13.02, N = 3SE +/- 67.90, N = 3SE +/- 36.96, N = 3118621983520179204111. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - 1000Mbit Objective - Parallel: 64Kernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC4K8K12K16K20KMin: 9837 / Avg: 11862.2 / Max: 20576Min: 19813 / Avg: 19834.67 / Max: 19858Min: 20080 / Avg: 20179 / Max: 20309Min: 20348 / Avg: 20411.33 / Max: 204761. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: TCP - Parallel: 1Kernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC5K10K15K20K25KSE +/- 109.87, N = 3SE +/- 212.79, N = 3SE +/- 273.77, N = 15SE +/- 85.32, N = 3128762189922637228221. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: TCP - Parallel: 1Kernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC4K8K12K16K20KMin: 12715 / Avg: 12876 / Max: 13086Min: 21503 / Avg: 21899 / Max: 22232Min: 22185 / Avg: 22636.67 / Max: 25667Min: 22717 / Avg: 22822 / Max: 229911. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Sched PipeKernel GCC 11.1.1Kernel GCCKernel LLVMKernel LLVM 11.150K100K150K200K250KSE +/- 296.37, N = 3SE +/- 569.18, N = 3SE +/- 896.38, N = 3SE +/- 861.60, N = 31990492099322142712174331. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Sched PipeKernel GCC 11.1.1Kernel GCCKernel LLVMKernel LLVM 11.140K80K120K160K200KMin: 198456 / Avg: 199048.67 / Max: 199353Min: 209020 / Avg: 209932 / Max: 210978Min: 212589 / Avg: 214271.33 / Max: 215649Min: 215718 / Avg: 217433 / Max: 2184361. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 2 - Type: ThreadKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC612182430SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.07, N = 3SE +/- 0.16, N = 323.5023.0621.9321.531. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 2 - Type: ThreadKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC510152025Min: 23.43 / Avg: 23.5 / Max: 23.55Min: 23.04 / Avg: 23.06 / Max: 23.08Min: 21.8 / Avg: 21.93 / Max: 22.06Min: 21.28 / Avg: 21.53 / Max: 21.811. (CC) gcc options: -lpthread

BenchmarkMutex

BenchmarkMutex is a mutex benchmark created by Malte Skarupke. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Semaphore Release And AcquireKernel GCCKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVM510152025SE +/- 0.07, N = 3SE +/- 0.00, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 319.119.019.019.01. (CXX) g++ options: -std=c++17 -lbenchmark -pthread
OpenBenchmarking.orgns, Fewer Is BetterBenchmarkMutexBenchmark: Semaphore Release And AcquireKernel GCCKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVM510152025Min: 19 / Avg: 19.07 / Max: 19.2Min: 19 / Avg: 19 / Max: 19Min: 18.9 / Avg: 19 / Max: 19.1Min: 19 / Avg: 19.03 / Max: 19.11. (CXX) g++ options: -std=c++17 -lbenchmark -pthread

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 1024 MiBKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC11002200330044005500SE +/- 18.98, N = 3SE +/- 4.69, N = 3SE +/- 7.65, N = 3SE +/- 1.90, N = 35152.915221.135252.545257.931. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 1024 MiBKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC9001800270036004500Min: 5123.74 / Avg: 5152.91 / Max: 5188.52Min: 5211.76 / Avg: 5221.13 / Max: 5225.98Min: 5240.36 / Avg: 5252.54 / Max: 5266.65Min: 5254.15 / Avg: 5257.93 / Max: 5260.091. (CC) gcc options: -O3 -march=native

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ThreadsKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.148121620SE +/- 0.68, N = 15SE +/- 0.64, N = 15SE +/- 0.70, N = 15SE +/- 0.04, N = 316.5516.0215.7514.57-lm-lm1. (CC) gcc options:
OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ThreadsKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.148121620Min: 14.86 / Avg: 16.55 / Max: 20.88Min: 14.67 / Avg: 16.02 / Max: 21.03Min: 14.03 / Avg: 15.75 / Max: 20.28Min: 14.52 / Avg: 14.57 / Max: 14.651. (CC) gcc options:

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 4096Kernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.1100K200K300K400K500KSE +/- 680.61, N = 3SE +/- 1305.06, N = 3SE +/- 1755.44, N = 3SE +/- 1315.59, N = 3448144457959473559474951
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 4096Kernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.180K160K240K320K400KMin: 446949 / Avg: 448144 / Max: 449306Min: 455926 / Avg: 457958.67 / Max: 460393Min: 470649 / Avg: 473558.67 / Max: 476715Min: 472334 / Avg: 474950.67 / Max: 476498

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 4096Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1100K200K300K400K500KSE +/- 3477.35, N = 3SE +/- 1532.35, N = 3SE +/- 4438.60, N = 3SE +/- 2203.58, N = 3468875471693479683483566
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 4096Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.180K160K240K320K400KMin: 463556 / Avg: 468875.33 / Max: 475415Min: 469441 / Avg: 471692.67 / Max: 474619Min: 475194 / Avg: 479683 / Max: 488560Min: 479190 / Avg: 483565.67 / Max: 486209

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 4096Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1110K220K330K440K550KSE +/- 2657.74, N = 3SE +/- 3051.56, N = 3SE +/- 1945.98, N = 3SE +/- 851.18, N = 3465636469668483511490914
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 4096Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.190K180K270K360K450KMin: 461645 / Avg: 465636 / Max: 470672Min: 463609 / Avg: 469668.33 / Max: 473330Min: 480531 / Avg: 483511 / Max: 487169Min: 489651 / Avg: 490914 / Max: 492534

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 1 - Type: ProcessKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.13691215SE +/- 0.026, N = 3SE +/- 0.028, N = 3SE +/- 0.230, N = 12SE +/- 0.083, N = 311.45810.38710.0849.5971. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 1 - Type: ProcessKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.13691215Min: 11.41 / Avg: 11.46 / Max: 11.5Min: 10.35 / Avg: 10.39 / Max: 10.44Min: 9.61 / Avg: 10.08 / Max: 11.74Min: 9.46 / Avg: 9.6 / Max: 9.741. (CC) gcc options: -lpthread

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 4096Kernel LLVMKernel GCCKernel GCC 11.1.1Kernel LLVM 11.1120K240K360K480K600KSE +/- 1970.04, N = 3SE +/- 4984.28, N = 3SE +/- 2367.23, N = 3SE +/- 1685.44, N = 3550089555085556704562576
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 4096Kernel LLVMKernel GCCKernel GCC 11.1.1Kernel LLVM 11.1100K200K300K400K500KMin: 547442 / Avg: 550089 / Max: 553940Min: 545317 / Avg: 555085 / Max: 561692Min: 552020 / Avg: 556704 / Max: 559643Min: 559207 / Avg: 562576 / Max: 564358

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 2048Kernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC120K240K360K480K600KSE +/- 1696.10, N = 3SE +/- 359.60, N = 3SE +/- 3911.99, N = 3SE +/- 1280.33, N = 3515245520300558725559665
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 2048Kernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC100K200K300K400K500KMin: 512096 / Avg: 515244.67 / Max: 517912Min: 519926 / Avg: 520300 / Max: 521019Min: 554753 / Avg: 558725.33 / Max: 566549Min: 557308 / Avg: 559665.33 / Max: 561710

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 512 MiBKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC8001600240032004000SE +/- 1.95, N = 3SE +/- 2.51, N = 3SE +/- 6.66, N = 3SE +/- 4.41, N = 33548.363665.783693.633719.911. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 512 MiBKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC6001200180024003000Min: 3544.5 / Avg: 3548.36 / Max: 3550.79Min: 3663.19 / Avg: 3665.78 / Max: 3670.8Min: 3685.93 / Avg: 3693.63 / Max: 3706.9Min: 3711.94 / Avg: 3719.91 / Max: 3727.151. (CC) gcc options: -O3 -march=native

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 512Kernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC300K600K900K1200K1500KSE +/- 7819.75, N = 12SE +/- 5290.86, N = 3SE +/- 1165.49, N = 3SE +/- 4109.25, N = 31120322113745311511751171213
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 512Kernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC200K400K600K800K1000KMin: 1059585 / Avg: 1120321.67 / Max: 1145681Min: 1128001 / Avg: 1137453 / Max: 1146299Min: 1149254 / Avg: 1151175.33 / Max: 1153279Min: 1166570 / Avg: 1171212.67 / Max: 1179407

iPerf

iPerf is a network bandwidth throughput testing software. This test profile is used for automated testing of an iperf client and requires you have access to an iperf server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - Parallel: 64Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.11530456075SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 367.167.167.167.11. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - Parallel: 64Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.11326395265Min: 67.1 / Avg: 67.1 / Max: 67.1Min: 67.1 / Avg: 67.1 / Max: 67.1Min: 67.1 / Avg: 67.1 / Max: 67.1Min: 67.1 / Avg: 67.1 / Max: 67.11. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - 100Mbit Objective - Parallel: 64Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.114002800420056007000SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 364006400640064001. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - 100Mbit Objective - Parallel: 64Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.111002200330044005500Min: 6400 / Avg: 6400 / Max: 6400Min: 6400 / Avg: 6400 / Max: 6400Min: 6400 / Avg: 6400 / Max: 6400Min: 6400 / Avg: 6400 / Max: 64001. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - Parallel: 32Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.1816243240SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 333.633.633.633.61. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - Parallel: 32Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.1714212835Min: 33.6 / Avg: 33.6 / Max: 33.6Min: 33.6 / Avg: 33.6 / Max: 33.6Min: 33.6 / Avg: 33.6 / Max: 33.6Min: 33.5 / Avg: 33.57 / Max: 33.61. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - 1000Mbit Objective - Parallel: 32Kernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC5K10K15K20K25KSE +/- 267.82, N = 3SE +/- 124.77, N = 3SE +/- 114.44, N = 3SE +/- 131.93, N = 3205042097421412214341. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - 1000Mbit Objective - Parallel: 32Kernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC4K8K12K16K20KMin: 19969 / Avg: 20503.67 / Max: 20799Min: 20832 / Avg: 20974.33 / Max: 21223Min: 21269 / Avg: 21411.67 / Max: 21638Min: 21177 / Avg: 21433.67 / Max: 216151. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - 100Mbit Objective - Parallel: 32Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.17001400210028003500SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 332003200320032001. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - 100Mbit Objective - Parallel: 32Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.16001200180024003000Min: 3200 / Avg: 3200 / Max: 3200Min: 3200 / Avg: 3200 / Max: 3200Min: 3200 / Avg: 3200 / Max: 3200Min: 3200 / Avg: 3200 / Max: 32001. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - Parallel: 1Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.10.23630.47260.70890.94521.1815SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31.051.051.051.051. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - Parallel: 1Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.1246810Min: 1.05 / Avg: 1.05 / Max: 1.05Min: 1.05 / Avg: 1.05 / Max: 1.05Min: 1.05 / Avg: 1.05 / Max: 1.05Min: 1.05 / Avg: 1.05 / Max: 1.051. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - 1000Mbit Objective - Parallel: 1Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.12004006008001000SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 310001000100010001. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - 1000Mbit Objective - Parallel: 1Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.12004006008001000Min: 1000 / Avg: 1000 / Max: 1000Min: 1000 / Avg: 1000 / Max: 1000Min: 1000 / Avg: 1000 / Max: 1000Min: 1000 / Avg: 1000 / Max: 10001. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - 100Mbit Objective - Parallel: 1Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.120406080100SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 31001001001001. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm
OpenBenchmarking.orgMbits/sec, More Is BetteriPerf 3.7Server Address: localhost - Server Port: 5201 - Duration: 10 Seconds - Test: UDP - 100Mbit Objective - Parallel: 1Kernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.120406080100Min: 100 / Avg: 100 / Max: 100Min: 100 / Avg: 100 / Max: 100Min: 100 / Avg: 100 / Max: 100Min: 100 / Avg: 100 / Max: 1001. (CC) gcc options: -O3 -march=native -lssl -lcrypto -lm

Hackbench

This is a benchmark of Hackbench, a test of the Linux kernel scheduler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 1 - Type: ThreadKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC3691215SE +/- 0.06, N = 3SE +/- 0.04, N = 3SE +/- 0.08, N = 3SE +/- 0.05, N = 312.1511.9811.3910.961. (CC) gcc options: -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterHackbenchCount: 1 - Type: ThreadKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC48121620Min: 12.08 / Avg: 12.15 / Max: 12.26Min: 11.94 / Avg: 11.98 / Max: 12.05Min: 11.28 / Avg: 11.39 / Max: 11.54Min: 10.86 / Avg: 10.96 / Max: 11.021. (CC) gcc options: -lpthread

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 2048Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1200K400K600K800K1000KSE +/- 1912.41, N = 3SE +/- 5943.58, N = 3SE +/- 1585.30, N = 3SE +/- 4963.31, N = 3807480808658811570815678
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 2048Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1140K280K420K560K700KMin: 804469 / Avg: 807479.67 / Max: 811028Min: 796783 / Avg: 808658 / Max: 815061Min: 809382 / Avg: 811569.67 / Max: 814651Min: 805878 / Avg: 815677.67 / Max: 821948

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 2048Kernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC200K400K600K800K1000KSE +/- 1812.61, N = 3SE +/- 4500.55, N = 3SE +/- 1382.62, N = 3SE +/- 856.11, N = 3809206812768814387819679
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 2048Kernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVMKernel GCC140K280K420K560K700KMin: 807324 / Avg: 809205.67 / Max: 812830Min: 803851 / Avg: 812768.33 / Max: 818288Min: 811833 / Avg: 814387 / Max: 816582Min: 818377 / Avg: 819679 / Max: 821293

t-test1

This is a test of t-test1 for basic memory allocator benchmarks. Note this test profile is currently very basic and the overall time does include the warmup time of the custom t-test1 compilation. Improvements welcome. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Bettert-test1 2017-01-13Threads: 2Kernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC3691215SE +/- 0.03, N = 3SE +/- 0.01, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 311.1411.0410.9610.941. (CC) gcc options: -pthread
OpenBenchmarking.orgSeconds, Fewer Is Bettert-test1 2017-01-13Threads: 2Kernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC3691215Min: 11.09 / Avg: 11.14 / Max: 11.18Min: 11.02 / Avg: 11.04 / Max: 11.07Min: 10.86 / Avg: 10.96 / Max: 11.02Min: 10.89 / Avg: 10.94 / Max: 11.041. (CC) gcc options: -pthread

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 256Kernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM500K1000K1500K2000K2500KSE +/- 13535.65, N = 3SE +/- 16570.75, N = 15SE +/- 22225.96, N = 7SE +/- 9721.39, N = 32467568250388925171332539979
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 256Kernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM400K800K1200K1600K2000KMin: 2442804 / Avg: 2467568 / Max: 2489421Min: 2322474 / Avg: 2503888.6 / Max: 2556699Min: 2392650 / Avg: 2517133 / Max: 2570436Min: 2524751 / Avg: 2539979.33 / Max: 2558062

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 2048Kernel GCC 11.1.1Kernel GCCKernel LLVM 11.1Kernel LLVM200K400K600K800K1000KSE +/- 8337.04, N = 3SE +/- 8681.93, N = 3SE +/- 4947.37, N = 3SE +/- 5079.92, N = 3941645944138948187950880
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 2048Kernel GCC 11.1.1Kernel GCCKernel LLVM 11.1Kernel LLVM160K320K480K640K800KMin: 925279 / Avg: 941644.67 / Max: 952592Min: 928842 / Avg: 944138.33 / Max: 958903Min: 938540 / Avg: 948187 / Max: 954916Min: 943673 / Avg: 950879.67 / Max: 960685

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 512 MiBKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC11002200330044005500SE +/- 12.30, N = 3SE +/- 10.80, N = 3SE +/- 9.52, N = 3SE +/- 8.07, N = 35146.675204.925233.325267.521. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 512 MiBKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC9001800270036004500Min: 5130.94 / Avg: 5146.67 / Max: 5170.92Min: 5190.72 / Avg: 5204.92 / Max: 5226.11Min: 5217.18 / Avg: 5233.32 / Max: 5250.14Min: 5252.3 / Avg: 5267.52 / Max: 5279.761. (CC) gcc options: -O3 -march=native

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 512Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1400K800K1200K1600K2000KSE +/- 1897.32, N = 3SE +/- 15480.54, N = 15SE +/- 16098.73, N = 3SE +/- 6345.89, N = 31772787179850918093301824207
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 512Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1300K600K900K1200K1500KMin: 1769044 / Avg: 1772786.67 / Max: 1775200Min: 1717444 / Avg: 1798508.73 / Max: 1862500Min: 1790832 / Avg: 1809330.33 / Max: 1841402Min: 1814847 / Avg: 1824206.67 / Max: 1836310

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 1024Kernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC200K400K600K800K1000KSE +/- 3059.81, N = 3SE +/- 3632.86, N = 3SE +/- 1701.21, N = 3SE +/- 4520.85, N = 3926277933478936841964065
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 1024Kernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC200K400K600K800K1000KMin: 921088 / Avg: 926277.33 / Max: 931681Min: 927316 / Avg: 933478.33 / Max: 939893Min: 934958 / Avg: 936841.33 / Max: 940237Min: 958879 / Avg: 964064.67 / Max: 973072

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Sequential FillKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC120K240K360K480K600KSE +/- 3481.21, N = 3SE +/- 3023.00, N = 3SE +/- 6929.94, N = 3SE +/- 2578.31, N = 34903545347705357965445721. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread
OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Sequential FillKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC90K180K270K360K450KMin: 483797 / Avg: 490354 / Max: 495660Min: 530026 / Avg: 534770 / Max: 540388Min: 522670 / Avg: 535796 / Max: 546213Min: 539675 / Avg: 544571.67 / Max: 5484201. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 256Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1300K600K900K1200K1500KSE +/- 4752.07, N = 3SE +/- 7325.27, N = 3SE +/- 1670.87, N = 3SE +/- 17761.76, N = 31204312120916112336881281031
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 256Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1200K400K600K800K1000KMin: 1196336 / Avg: 1204312.33 / Max: 1212776Min: 1195216 / Avg: 1209161.33 / Max: 1220023Min: 1231141 / Avg: 1233688 / Max: 1236835Min: 1245534 / Avg: 1281031 / Max: 1299968

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 1024Kernel GCCKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1300K600K900K1200K1500KSE +/- 2423.58, N = 3SE +/- 3612.27, N = 3SE +/- 2398.52, N = 3SE +/- 3846.75, N = 31421362144240714538271471645
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 1024Kernel GCCKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1300K600K900K1200K1500KMin: 1417732 / Avg: 1421362.33 / Max: 1425959Min: 1437526 / Avg: 1442406.67 / Max: 1449460Min: 1450095 / Avg: 1453826.67 / Max: 1458303Min: 1463961 / Avg: 1471645 / Max: 1475818

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 1024Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1300K600K900K1200K1500KSE +/- 2998.97, N = 3SE +/- 2437.79, N = 3SE +/- 1557.87, N = 3SE +/- 6078.06, N = 31280989128111212972161324519
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 1024Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1200K400K600K800K1000KMin: 1276799 / Avg: 1280989.33 / Max: 1286801Min: 1278538 / Avg: 1281112 / Max: 1285985Min: 1294175 / Avg: 1297216 / Max: 1299324Min: 1314858 / Avg: 1324518.67 / Max: 1335739

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 1024Kernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.1300K600K900K1200K1500KSE +/- 16798.04, N = 3SE +/- 3654.53, N = 3SE +/- 9574.67, N = 3SE +/- 5287.17, N = 31279652128548713169131322294
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 1024Kernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.1200K400K600K800K1000KMin: 1256042 / Avg: 1279652 / Max: 1312156Min: 1278282 / Avg: 1285486.67 / Max: 1290155Min: 1298638 / Avg: 1316913.33 / Max: 1331004Min: 1311939 / Avg: 1322294 / Max: 1329327

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Sequential FillKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.148121620SE +/- 0.05, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 3SE +/- 0.10, N = 313.9613.7313.5513.41-lsqlite3-lsqlite31. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Sequential FillKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.148121620Min: 13.87 / Avg: 13.96 / Max: 14.06Min: 13.63 / Avg: 13.73 / Max: 13.82Min: 13.48 / Avg: 13.55 / Max: 13.59Min: 13.27 / Avg: 13.41 / Max: 13.611. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Sequential FillKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1816243240SE +/- 0.12, N = 3SE +/- 0.12, N = 3SE +/- 0.10, N = 3SE +/- 0.27, N = 331.632.232.632.9-lsqlite3-lsqlite31. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Sequential FillKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1714212835Min: 31.4 / Avg: 31.6 / Max: 31.8Min: 32 / Avg: 32.2 / Max: 32.4Min: 32.5 / Avg: 32.6 / Max: 32.8Min: 32.4 / Avg: 32.93 / Max: 33.31. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random DeleteKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.13691215SE +/- 0.03, N = 3SE +/- 0.15, N = 3SE +/- 0.07, N = 3SE +/- 0.11, N = 313.4613.4513.1612.85-lsqlite3-lsqlite31. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random DeleteKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.148121620Min: 13.42 / Avg: 13.46 / Max: 13.53Min: 13.24 / Avg: 13.45 / Max: 13.74Min: 13.06 / Avg: 13.16 / Max: 13.31Min: 12.73 / Avg: 12.85 / Max: 13.061. (CXX) g++ options: -O3 -lsnappy -lpthread

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 128Kernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.1300K600K900K1200K1500KSE +/- 5815.34, N = 3SE +/- 4775.86, N = 3SE +/- 4747.41, N = 3SE +/- 4904.13, N = 31336817136356113747451411134
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Unix Domain Socket - Message Bytes: 128Kernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.1200K400K600K800K1000KMin: 1325866 / Avg: 1336817.33 / Max: 1345685Min: 1357779 / Avg: 1363560.67 / Max: 1373036Min: 1365411 / Avg: 1374744.67 / Max: 1380920Min: 1403961 / Avg: 1411134.33 / Max: 1420514

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 128Kernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC700K1400K2100K2800K3500KSE +/- 11315.58, N = 3SE +/- 22099.44, N = 3SE +/- 33830.67, N = 15SE +/- 34085.75, N = 42722030289172730340983040261
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 128Kernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC500K1000K1500K2000K2500KMin: 2708003 / Avg: 2722030 / Max: 2744424Min: 2853697 / Avg: 2891727 / Max: 2930247Min: 2874883 / Avg: 3034097.53 / Max: 3331807Min: 2973777 / Avg: 3040260.5 / Max: 3132091

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 128Kernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM600K1200K1800K2400K3000KSE +/- 1162.56, N = 3SE +/- 41324.96, N = 3SE +/- 24608.70, N = 9SE +/- 10452.71, N = 32906209294322729713432985001
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 128Kernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM500K1000K1500K2000K2500KMin: 2904682 / Avg: 2906209 / Max: 2908491Min: 2860857 / Avg: 2943226.67 / Max: 2990301Min: 2823848 / Avg: 2971343 / Max: 3039418Min: 2966034 / Avg: 2985001.33 / Max: 3002098

perf-bench

This test profile is used for running Linux perf-bench, the benchmark support within the Linux kernel's perf tool. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Syscall BasicKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC4M8M12M16M20MSE +/- 12906.55, N = 3SE +/- 44903.68, N = 3SE +/- 101385.04, N = 3SE +/- 6821.55, N = 3165862831662756616699690169079391. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma
OpenBenchmarking.orgops/sec, More Is Betterperf-benchBenchmark: Syscall BasicKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC3M6M9M12M15MMin: 16561188 / Avg: 16586283.33 / Max: 16604066Min: 16538419 / Avg: 16627566.33 / Max: 16681552Min: 16496921 / Avg: 16699690.33 / Max: 16801552Min: 16894837 / Avg: 16907938.67 / Max: 169177851. (CC) gcc options: -O6 -ggdb3 -funwind-tables -std=gnu99 -lunwind-x86_64 -lunwind -llzma -Xlinker -export-dynamic -lpthread -lrt -lm -ldl -lelf -ldebuginfod -ldw -lcrypto -lslang -lpython2.7 -lutil -lbfd -lopcodes -liberty -lz -lzstd -lcap -lnuma

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 512Kernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM400K800K1200K1600K2000KSE +/- 3867.40, N = 3SE +/- 21799.46, N = 4SE +/- 2605.04, N = 3SE +/- 11997.79, N = 32016930204467720540862062968
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: TCP Socket - Message Bytes: 512Kernel GCC 11.1.1Kernel LLVM 11.1Kernel GCCKernel LLVM400K800K1200K1600K2000KMin: 2012002 / Avg: 2016930 / Max: 2024557Min: 1983821 / Avg: 2044677.25 / Max: 2076695Min: 2049426 / Avg: 2054085.67 / Max: 2058434Min: 2039006 / Avg: 2062967.67 / Max: 2076053

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create FilesKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC918273645SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.24, N = 3SE +/- 0.04, N = 339.4039.0038.8236.75-lm-lm1. (CC) gcc options:
OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create FilesKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC816243240Min: 39.31 / Avg: 39.4 / Max: 39.49Min: 38.92 / Avg: 39 / Max: 39.1Min: 38.51 / Avg: 38.82 / Max: 39.29Min: 36.69 / Avg: 36.75 / Max: 36.831. (CC) gcc options:

OpenBenchmarking.orgNs Per Event, Fewer Is BetterOSBenchTest: Memory AllocationsKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC306090120150SE +/- 1.67, N = 3SE +/- 0.44, N = 3SE +/- 0.34, N = 3SE +/- 0.11, N = 3143.74141.11137.38135.70-lm-lm1. (CC) gcc options:
OpenBenchmarking.orgNs Per Event, Fewer Is BetterOSBenchTest: Memory AllocationsKernel GCC 11.1.1Kernel LLVMKernel LLVM 11.1Kernel GCC306090120150Min: 141.93 / Avg: 143.74 / Max: 147.07Min: 140.25 / Avg: 141.11 / Max: 141.7Min: 136.82 / Avg: 137.38 / Max: 137.98Min: 135.49 / Avg: 135.7 / Max: 135.881. (CC) gcc options:

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 512Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1400K800K1200K1600K2000KSE +/- 24548.29, N = 3SE +/- 8871.35, N = 3SE +/- 19241.60, N = 3SE +/- 14438.74, N = 31746678174710317763331835701
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 512Kernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1300K600K900K1200K1500KMin: 1698413 / Avg: 1746678 / Max: 1778603Min: 1729647 / Avg: 1747103.33 / Max: 1758581Min: 1738205 / Avg: 1776333.33 / Max: 1799913Min: 1806852 / Avg: 1835700.67 / Max: 1851242

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Launch ProgramsKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC20406080100SE +/- 0.64, N = 3SE +/- 0.20, N = 3SE +/- 0.41, N = 3SE +/- 0.19, N = 397.4395.5193.3291.06-lm-lm1. (CC) gcc options:
OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Launch ProgramsKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC20406080100Min: 96.21 / Avg: 97.43 / Max: 98.38Min: 95.28 / Avg: 95.51 / Max: 95.9Min: 92.58 / Avg: 93.32 / Max: 94.01Min: 90.73 / Avg: 91.06 / Max: 91.41. (CC) gcc options:

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ProcessesKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1918273645SE +/- 0.18, N = 3SE +/- 0.16, N = 3SE +/- 0.28, N = 3SE +/- 0.48, N = 340.9340.3139.4939.48-lm-lm1. (CC) gcc options:
OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create ProcessesKernel LLVMKernel LLVM 11.1Kernel GCCKernel GCC 11.1.1918273645Min: 40.57 / Avg: 40.93 / Max: 41.12Min: 40.05 / Avg: 40.31 / Max: 40.61Min: 39.05 / Avg: 39.49 / Max: 40.02Min: 38.58 / Avg: 39.48 / Max: 40.21. (CC) gcc options:

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random ReadKernel GCCKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVM0.54251.0851.62752.172.7125SE +/- 0.019, N = 9SE +/- 0.012, N = 3SE +/- 0.021, N = 7SE +/- 0.027, N = 42.4112.4022.3532.293-lsqlite3-lsqlite31. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random ReadKernel GCCKernel GCC 11.1.1Kernel LLVM 11.1Kernel LLVM246810Min: 2.29 / Avg: 2.41 / Max: 2.49Min: 2.38 / Avg: 2.4 / Max: 2.42Min: 2.29 / Avg: 2.35 / Max: 2.46Min: 2.24 / Avg: 2.29 / Max: 2.361. (CXX) g++ options: -O3 -lsnappy -lpthread

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 256Kernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.1500K1000K1500K2000K2500KSE +/- 4551.71, N = 3SE +/- 5587.94, N = 3SE +/- 4874.11, N = 3SE +/- 18030.45, N = 32150341227845422997032306298
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 256Kernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.1400K800K1200K1600K2000KMin: 2141241 / Avg: 2150340.67 / Max: 2155117Min: 2267504 / Avg: 2278453.67 / Max: 2285866Min: 2291295 / Avg: 2299703.33 / Max: 2308179Min: 2270240 / Avg: 2306298.33 / Max: 2324700

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 256Kernel LLVM 11.1Kernel GCCKernel LLVMKernel GCC 11.1.1500K1000K1500K2000K2500KSE +/- 18436.29, N = 3SE +/- 17228.51, N = 3SE +/- 31763.45, N = 3SE +/- 27586.82, N = 32283738233737723517612360107
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: Unnamed Pipe - Message Bytes: 256Kernel LLVM 11.1Kernel GCCKernel LLVMKernel GCC 11.1.1400K800K1200K1600K2000KMin: 2254871 / Avg: 2283738.33 / Max: 2318039Min: 2303457 / Avg: 2337376.67 / Max: 2359586Min: 2290971 / Avg: 2351761.33 / Max: 2398130Min: 2315034 / Avg: 2360107 / Max: 2410201

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 128 MiBKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC8001600240032004000SE +/- 3.01, N = 3SE +/- 6.48, N = 3SE +/- 1.33, N = 3SE +/- 2.94, N = 33623.863627.973648.543715.491. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy, Fixed Block Size - Array Size: 128 MiBKernel LLVMKernel GCC 11.1.1Kernel LLVM 11.1Kernel GCC6001200180024003000Min: 3617.9 / Avg: 3623.86 / Max: 3627.5Min: 3618.76 / Avg: 3627.97 / Max: 3640.46Min: 3645.91 / Avg: 3648.54 / Max: 3650.21Min: 3711.98 / Avg: 3715.49 / Max: 3721.341. (CC) gcc options: -O3 -march=native

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Seek RandomKernel LLVM 11.1Kernel GCCKernel LLVMKernel GCC 11.1.10.76771.53542.30313.07083.8385SE +/- 0.028, N = 3SE +/- 0.029, N = 3SE +/- 0.029, N = 3SE +/- 0.047, N = 33.4123.3933.3593.358-lsqlite3-lsqlite31. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Seek RandomKernel LLVM 11.1Kernel GCCKernel LLVMKernel GCC 11.1.1246810Min: 3.36 / Avg: 3.41 / Max: 3.45Min: 3.34 / Avg: 3.39 / Max: 3.43Min: 3.31 / Avg: 3.36 / Max: 3.4Min: 3.27 / Avg: 3.36 / Max: 3.421. (CXX) g++ options: -O3 -lsnappy -lpthread

IPC_benchmark

IPC_benchmark is a Linux inter-process communication benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 128Kernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC700K1400K2100K2800K3500KSE +/- 35347.02, N = 3SE +/- 18345.59, N = 3SE +/- 8055.70, N = 3SE +/- 29770.09, N = 32649855280115729869123043686
OpenBenchmarking.orgMessages Per Second, More Is BetterIPC_benchmarkType: FIFO Named Pipe - Message Bytes: 128Kernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC500K1000K1500K2000K2500KMin: 2588751 / Avg: 2649855.33 / Max: 2711196Min: 2766827 / Avg: 2801156.67 / Max: 2829537Min: 2970955 / Avg: 2986912.33 / Max: 2996816Min: 2984255 / Avg: 3043685.67 / Max: 3076527

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random FillKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC48121620SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.14, N = 14SE +/- 0.11, N = 314.7214.4814.1614.14-lsqlite3-lsqlite31. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random FillKernel LLVM 11.1Kernel LLVMKernel GCC 11.1.1Kernel GCC48121620Min: 14.59 / Avg: 14.72 / Max: 14.79Min: 14.42 / Avg: 14.48 / Max: 14.6Min: 13.74 / Avg: 14.16 / Max: 15.78Min: 13.93 / Avg: 14.14 / Max: 14.291. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Random FillKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.1714212835SE +/- 0.12, N = 3SE +/- 0.13, N = 3SE +/- 0.24, N = 3SE +/- 0.30, N = 1430.030.531.231.2-lsqlite3-lsqlite31. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Random FillKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.1714212835Min: 29.8 / Avg: 29.97 / Max: 30.2Min: 30.2 / Avg: 30.47 / Max: 30.6Min: 30.9 / Avg: 31.23 / Max: 31.7Min: 28 / Avg: 31.24 / Max: 32.21. (CXX) g++ options: -O3 -lsnappy -lpthread

MBW

This is a basic/simple memory (RAM) bandwidth benchmark for memory copy operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 128 MiBKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC11002200330044005500SE +/- 19.56, N = 3SE +/- 6.49, N = 3SE +/- 12.43, N = 3SE +/- 7.70, N = 35160.395171.815172.975259.711. (CC) gcc options: -O3 -march=native
OpenBenchmarking.orgMiB/s, More Is BetterMBW 2018-09-08Test: Memory Copy - Array Size: 128 MiBKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC9001800270036004500Min: 5121.45 / Avg: 5160.39 / Max: 5183.02Min: 5163.91 / Avg: 5171.81 / Max: 5184.67Min: 5155.83 / Avg: 5172.97 / Max: 5197.15Min: 5249.85 / Avg: 5259.71 / Max: 5274.91. (CC) gcc options: -O3 -march=native

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Hot ReadKernel GCC 11.1.1Kernel GCCKernel LLVM 11.1Kernel LLVM0.53841.07681.61522.15362.692SE +/- 0.020, N = 3SE +/- 0.023, N = 3SE +/- 0.027, N = 3SE +/- 0.006, N = 32.3932.3892.3172.310-lsqlite3-lsqlite31. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Hot ReadKernel GCC 11.1.1Kernel GCCKernel LLVM 11.1Kernel LLVM246810Min: 2.36 / Avg: 2.39 / Max: 2.42Min: 2.34 / Avg: 2.39 / Max: 2.42Min: 2.27 / Avg: 2.32 / Max: 2.36Min: 2.3 / Avg: 2.31 / Max: 2.321. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: OverwriteKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.148121620SE +/- 0.09, N = 3SE +/- 0.09, N = 3SE +/- 0.12, N = 3SE +/- 0.06, N = 314.7314.5714.1313.92-lsqlite3-lsqlite31. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: OverwriteKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.148121620Min: 14.62 / Avg: 14.73 / Max: 14.92Min: 14.48 / Avg: 14.57 / Max: 14.74Min: 13.93 / Avg: 14.12 / Max: 14.33Min: 13.84 / Avg: 13.92 / Max: 14.041. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: OverwriteKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.1714212835SE +/- 0.17, N = 3SE +/- 0.19, N = 3SE +/- 0.23, N = 3SE +/- 0.19, N = 329.930.331.231.7-lsqlite3-lsqlite31. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: OverwriteKernel LLVM 11.1Kernel LLVMKernel GCCKernel GCC 11.1.1714212835Min: 29.6 / Avg: 29.93 / Max: 30.1Min: 29.9 / Avg: 30.27 / Max: 30.5Min: 30.8 / Avg: 31.2 / Max: 31.6Min: 31.3 / Avg: 31.67 / Max: 31.91. (CXX) g++ options: -O3 -lsnappy -lpthread

ctx_clock

Ctx_clock is a simple test program to measure the context switch time in clock cycles. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgClocks, Fewer Is Betterctx_clockContext Switch TimeKernel GCC 11.1.1Kernel GCCKernel LLVM 11.1Kernel LLVM60120180240300SE +/- 0.88, N = 3SE +/- 0.00, N = 3SE +/- 0.33, N = 3SE +/- 0.33, N = 3268266264264
OpenBenchmarking.orgClocks, Fewer Is Betterctx_clockContext Switch TimeKernel GCC 11.1.1Kernel GCCKernel LLVM 11.1Kernel LLVM50100150200250Min: 266 / Avg: 267.67 / Max: 269Min: 266 / Avg: 266 / Max: 266Min: 264 / Avg: 264.33 / Max: 265Min: 264 / Avg: 264.33 / Max: 265

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Fill SyncKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC48121620SE +/- 0.19, N = 3SE +/- 0.06, N = 3SE +/- 0.15, N = 3SE +/- 0.26, N = 418.0617.6817.6817.44-lsqlite3-lsqlite31. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Fill SyncKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC510152025Min: 17.68 / Avg: 18.06 / Max: 18.29Min: 17.59 / Avg: 17.68 / Max: 17.8Min: 17.49 / Avg: 17.68 / Max: 17.97Min: 16.72 / Avg: 17.43 / Max: 17.971. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Fill SyncKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC612182430SE +/- 0.26, N = 3SE +/- 0.12, N = 3SE +/- 0.21, N = 3SE +/- 0.31, N = 424.324.824.925.1-lsqlite3-lsqlite31. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Fill SyncKernel LLVM 11.1Kernel GCC 11.1.1Kernel LLVMKernel GCC612182430Min: 23.9 / Avg: 24.3 / Max: 24.8Min: 24.6 / Avg: 24.83 / Max: 25Min: 24.5 / Avg: 24.9 / Max: 25.2Min: 24.3 / Avg: 25.05 / Max: 25.81. (CXX) g++ options: -O3 -lsnappy -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: NUMAKernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.1918273645SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3SE +/- 0.00, N = 3383838381. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc
OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: NUMAKernel LLVMKernel GCCKernel LLVM 11.1Kernel GCC 11.1.1816243240Min: 38 / Avg: 38 / Max: 38Min: 38 / Avg: 38 / Max: 38Min: 38 / Avg: 38 / Max: 38Min: 38 / Avg: 38 / Max: 381. (CC) gcc options: -O2 -std=gnu99 -lm -laio -lbsd -lcrypt -lrt -lz -ldl -lpthread -lc

146 Results Shown

Hackbench
WireGuard + Linux Networking Stack Stress Test
Hackbench
Tinymembench:
  Standard Memset
  Standard Memcpy
Hackbench
BenchmarkMutex
pmbench
Facebook RocksDB
Hackbench
Stress-NG
SQLite Speedtest
pmbench:
  4 - 80% Reads 20% Writes
  2 - 100% Reads
Stress-NG
MBW
Stress-NG
Schbench
Hackbench
MBW
pmbench
BenchmarkMutex:
  Mutex Lock Unlock pthread_mutex
  Shared Mutex Lock Shared
perf-bench
PostMark
pmbench:
  2 - 50%
  2 - 100% Writes
perf-bench
pmbench:
  2 - 80% Reads 20% Writes
  1 - 50%
  4 - 50%
  1 - 100% Writes
  4 - 100% Writes
  4 - 100% Reads
Facebook RocksDB:
  Rand Fill
  Rand Fill Sync
  Rand Read
Apache Benchmark
perf-bench
Stress-NG
BenchmarkMutex
PostgreSQL pgbench:
  100 - 1 - Read Only - Average Latency
  100 - 1 - Read Only
Hackbench:
  4 - Thread
  2 - Process
  4 - Process
BenchmarkMutex
Stress-NG:
  SENDFILE
  Glibc Qsort Data Sorting
t-test1
perf-bench
iPerf
Stress-NG:
  CPU Stress
  Malloc
  Crypto
  MEMFD
perf-bench
Stress-NG:
  Memory Copying
  Socket Activity
  Glibc C String Functions
  Context Switching
  Vector Math
  Matrix Math
  Forking
  Atomic
MBW
BenchmarkMutex:
  Mutex Lock Unlock std::mutex
  Mutex Lock Unlock ticket_spinlock
PostgreSQL pgbench:
  1 - 250 - Read Only - Average Latency
  1 - 250 - Read Only
  1 - 1 - Read Write - Average Latency
  1 - 1 - Read Write
  1 - 100 - Read Only - Average Latency
  1 - 100 - Read Only
  1 - 50 - Read Only - Average Latency
  1 - 50 - Read Only
  1 - 1 - Read Only - Average Latency
  1 - 1 - Read Only
iPerf
OpenSSL
iPerf:
  5201 - 10 Seconds - UDP - 1000Mbit Objective - 64
  5201 - 10 Seconds - TCP - 1
perf-bench
Hackbench
BenchmarkMutex
MBW
OSBench
IPC_benchmark:
  Unnamed Unix Domain Socket - 4096
  Unnamed Pipe - 4096
  FIFO Named Pipe - 4096
Hackbench
IPC_benchmark:
  TCP Socket - 4096
  Unnamed Unix Domain Socket - 2048
MBW
IPC_benchmark
iPerf:
  5201 - 10 Seconds - UDP - 64
  5201 - 10 Seconds - UDP - 100Mbit Objective - 64
  5201 - 10 Seconds - UDP - 32
  5201 - 10 Seconds - UDP - 1000Mbit Objective - 32
  5201 - 10 Seconds - UDP - 100Mbit Objective - 32
  5201 - 10 Seconds - UDP - 1
  5201 - 10 Seconds - UDP - 1000Mbit Objective - 1
  5201 - 10 Seconds - UDP - 100Mbit Objective - 1
Hackbench
IPC_benchmark:
  FIFO Named Pipe - 2048
  Unnamed Pipe - 2048
t-test1
IPC_benchmark:
  TCP Socket - 256
  TCP Socket - 2048
MBW
IPC_benchmark:
  Unnamed Pipe - 512
  Unnamed Unix Domain Socket - 1024
Facebook RocksDB
IPC_benchmark:
  Unnamed Unix Domain Socket - 256
  TCP Socket - 1024
  FIFO Named Pipe - 1024
  Unnamed Pipe - 1024
LevelDB:
  Seq Fill:
    Microseconds Per Op
    MB/s
  Rand Delete:
    Microseconds Per Op
IPC_benchmark:
  Unnamed Unix Domain Socket - 128
  Unnamed Pipe - 128
  TCP Socket - 128
perf-bench
IPC_benchmark
OSBench:
  Create Files
  Memory Allocations
IPC_benchmark
OSBench:
  Launch Programs
  Create Processes
LevelDB
IPC_benchmark:
  FIFO Named Pipe - 256
  Unnamed Pipe - 256
MBW
LevelDB
IPC_benchmark
LevelDB:
  Rand Fill:
    Microseconds Per Op
    MB/s
MBW
LevelDB:
  Hot Read
  Overwrite
  Overwrite
ctx_clock
LevelDB:
  Fill Sync:
    Microseconds Per Op
    MB/s
Stress-NG

Kernel LLVM

Processor: AMD Phenom II X4 965 @ 3.30GHz (4 Cores), Motherboard: MSI 890FXA-GD70 (MS-7640) v1.0 (V1.8 BIOS), Chipset: AMD RD890 PCI-e GFX Hydra part, Memory: 16GB, Disk: 116GB Samsung SSD 840 + 200GB Seagate ST3200822AS, Graphics: MSI NVIDIA GeForce GTX 650 Ti 1GB, Audio: Realtek ALC892, Network: Realtek RTL8111/8168/8411

OS: ManjaroLinux 21.0.4, Kernel: 5.12.2-153-tkg-pds-llvm (x86_64), Desktop: KDE Plasma 5.21.4, Display Server: X Server 1.20.11, Display Driver: NVIDIA 465.27, OpenGL: 4.6.0, Vulkan: 1.2.168, Compiler: GCC 10.2.0 + Clang 13.0.0, File-System: btrfs, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: always
Environment Notes: __GL_ExperimentalPerfStrategy=1 __GL_IGNORE_GLSL_EXT_REQS=1 __GL_THREADED_OPTIMIZATIONS=1 __GL_SHADER_DISK_CACHE=1 __GL_SHADER_CACHE=1 __GL_ALLOW_UNOFFICIAL_PROTOCOL=1
Disk Notes: BFQ / commit=120,compress=zstd:3,discard=async,lazytime,noatime,rw,space_cache,ssd_spread,subvol=/@home,subvolid=258 / Block Size: 4096
Processor Notes: Scaling Governor: acpi-cpufreq performance - CPU Microcode: 0x10000c8
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Not affected + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable STIBP: disabled + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 9 May 2021 18:32 by user linux.

Kernel GCC

Processor: AMD Phenom II X4 965 @ 3.30GHz (4 Cores), Motherboard: MSI 890FXA-GD70 (MS-7640) v1.0 (V1.8 BIOS), Chipset: AMD RD890 PCI-e GFX Hydra part, Memory: 16GB, Disk: 116GB Samsung SSD 840 + 200GB Seagate ST3200822AS, Graphics: MSI NVIDIA GeForce GTX 650 Ti 1GB, Audio: Realtek ALC892, Network: Realtek RTL8111/8168/8411

OS: ManjaroLinux 21.0.4, Kernel: 5.12.2-153-tkg-pds (x86_64), Desktop: KDE Plasma 5.21.4, Display Server: X Server 1.20.11, Display Driver: NVIDIA 465.27, OpenGL: 4.6.0, Vulkan: 1.2.168, Compiler: GCC 10.2.0 + Clang 13.0.0, File-System: btrfs, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: always
Environment Notes: __GL_ExperimentalPerfStrategy=1 __GL_IGNORE_GLSL_EXT_REQS=1 __GL_THREADED_OPTIMIZATIONS=1 __GL_SHADER_DISK_CACHE=1 __GL_SHADER_CACHE=1 __GL_ALLOW_UNOFFICIAL_PROTOCOL=1
Disk Notes: BFQ / commit=120,compress=zstd:3,discard=async,lazytime,noatime,rw,space_cache,ssd_spread,subvol=/@home,subvolid=258 / Block Size: 4096
Processor Notes: Scaling Governor: acpi-cpufreq performance - CPU Microcode: 0x10000c8
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Not affected + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable STIBP: disabled + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 10 May 2021 07:33 by user linux.

Kernel LLVM 11.1

Processor: AMD Phenom II X4 965 @ 3.30GHz (4 Cores), Motherboard: MSI 890FXA-GD70 (MS-7640) v1.0 (V1.8 BIOS), Chipset: AMD RD890 PCI-e GFX Hydra part, Memory: 16GB, Disk: 116GB Samsung SSD 840 + 200GB Seagate ST3200822AS, Graphics: MSI NVIDIA GeForce GTX 650 Ti 1GB, Audio: Realtek ALC892, Network: Realtek RTL8111/8168/8411

OS: ManjaroLinux 21.0.4, Kernel: 5.12.2-153-tkg-pds-llvm (x86_64), Desktop: KDE Plasma 5.21.4, Display Server: X Server 1.20.11, Display Driver: NVIDIA 465.27, OpenGL: 4.6.0, Vulkan: 1.2.168, Compiler: GCC 10.2.0 + Clang 11.1.0 + LLVM 11.1.0, File-System: btrfs, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: always
Environment Notes: __GL_ExperimentalPerfStrategy=1 __GL_IGNORE_GLSL_EXT_REQS=1 __GL_THREADED_OPTIMIZATIONS=1 __GL_SHADER_DISK_CACHE=1 __GL_SHADER_CACHE=1 __GL_ALLOW_UNOFFICIAL_PROTOCOL=1
Disk Notes: BFQ / commit=120,compress=zstd:3,discard=async,lazytime,noatime,rw,space_cache,ssd_spread,subvol=/@home,subvolid=258 / Block Size: 4096
Processor Notes: Scaling Governor: acpi-cpufreq performance - CPU Microcode: 0x10000c8
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Not affected + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable STIBP: disabled + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 10 May 2021 15:12 by user linux.

Kernel GCC 11.1.1

Processor: AMD Phenom II X4 965 @ 3.30GHz (4 Cores), Motherboard: MSI 890FXA-GD70 (MS-7640) v1.0 (V1.8 BIOS), Chipset: AMD RD890 PCI-e GFX Hydra part, Memory: 16GB, Disk: 116GB Samsung SSD 840 + 200GB Seagate ST3200822AS, Graphics: MSI NVIDIA GeForce GTX 650 Ti 1GB, Audio: Realtek ALC892, Network: Realtek RTL8111/8168/8411

OS: ManjaroLinux 21.0.4, Kernel: 5.12.2-154-tkg-pds (x86_64), Desktop: KDE Plasma 5.21.4, Display Server: X Server 1.20.11, Display Driver: NVIDIA 465.27, OpenGL: 4.6.0, Vulkan: 1.2.168, Compiler: GCC 10.2.0 + Clang 11.1.0 + LLVM 11.1.0, File-System: btrfs, Screen Resolution: 1920x1080

Kernel Notes: Transparent Huge Pages: always
Environment Notes: __GL_ExperimentalPerfStrategy=1 LTO_EXTRA_CFLAGS="-fgraphite-identity -floop-nest-optimize -fgcse-las -fgcse-sm -floop-interchange -floop-strip-mine -floop-block -ftree-loop-distribution -fno-signed-zeros -fno-trapping-math -fassociative-math -freciprocal-math -fno-math-errno -ffinite-math-only -fno-rounding-math -fno-signaling-nans -fcx-limited-range -fexcess-precision=fast -fdevirtualize-at-ltrans -feliminate-unused-debug-types -ftree-vectorize -mfpmath=sse -fno-semantic-interposition -ftree-loop-distribute-patterns -malign-data=cacheline -ftree-loop-vectorize -fbranch-target-load-optimize -fno-common" __GL_IGNORE_GLSL_EXT_REQS=1 __GL_THREADED_OPTIMIZATIONS=1 __GL_SHADER_DISK_CACHE=1 __GL_SHADER_CACHE=1 __GL_ALLOW_UNOFFICIAL_PROTOCOL=1
Disk Notes: BFQ / commit=120,compress=zstd:3,discard=async,lazytime,noatime,rw,space_cache,ssd_spread,subvol=/@home,subvolid=258 / Block Size: 4096
Processor Notes: Scaling Governor: acpi-cpufreq performance - CPU Microcode: 0x10000c8
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Not affected + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable STIBP: disabled + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 12 May 2021 17:20 by user linux.